"Fossies" - the Fresh Open Source Software Archive

Member "glusterfs-8.6/ChangeLog" (20 Aug 2021, 10080187 Bytes) of package /linux/misc/glusterfs-8.6.tar.gz:


As a special service "Fossies" has tried to format the requested text file into HTML format (style: standard) with prefixed line numbers. Alternatively you can here view or download the uninterpreted source code file.

    1 ===== git log ====
    2 commit 88dc7324f2e862921abbf901e341a69375fcc13a
    3 Author: Shwetha Acharya <sacharya@redhat.com>
    4 Date:   Thu Aug 19 21:29:05 2021 +0530
    5 
    6     doc: Added release 8.6 notes (#2718)
    7     
    8     * doc: Added release 8.6 notes
    9     
   10     Change-Id: I4efcb4351433fa785ae5766c9289871a06d64a32
   11     Updates: #2702
   12     Signed-off-by: Shwetha K Acharya <sacharya@redhat.com>
   13     
   14     * Release notes: remove blank line
   15     
   16     * Remove blank line again
   17 
   18 commit 263b262d836a07d56a2f04d287ca4b9fad80bd8d
   19 Author: Xavi Hernandez <xhernandez@redhat.com>
   20 Date:   Tue Aug 17 09:17:57 2021 +0200
   21 
   22     geo-rep: Improve handling of gfid mismatches (#2422)
   23     
   24     In some circumstances geo-replication can detect mismatching gfids
   25     between primary and secondary. These entries are fixed in an iterative
   26     way, assuming that after a fix, a previously failing entry could
   27     succeed.
   28     
   29     Previous code was trying to fix them in a loop that can be executed
   30     up to 10 times. If some entry cannot be fixed after 10 attempts, it's
   31     discarded. These fixes are very slow, so trying to do them many times
   32     causes geo-replication to get out of sync.
   33     
   34     To minimize the number of iterations done, this patch checks if the
   35     number of entries and failures remains constant after each iteration.
   36     If they are constant, it means that nothing else can be fixed, so it
   37     makes no sense to do more iterations. This reduces the number of
   38     iterations to 2 or 3 in most of the cases, improving geo-replication
   39     performance.
   40     
   41     Fixes: #2388
   42     Change-Id: I6d9a623a60045694e1a832195e1dc1fb9e88ae54
   43     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
   44 
   45 commit 462721efddbd6e437031a1fecf0f6713fbbb1675
   46 Author: Nikhil Ladha <nladha@redhat.com>
   47 Date:   Tue Aug 10 11:52:27 2021 +0530
   48 
   49     glusterd: reset mgmt_v3_lock_timeout after it be used (#2689)
   50     
   51     Problem:
   52     in mgmt v3, lock will be released after 3 minutes by default.
   53     for those commands which will take more time, it can recieve
   54     a timeout parameter instead of the default 3 minutes.
   55     but commit  91cbcd1 broke this method, reset timeout before it be used.
   56     so that lock release timeout is always 3 minutes.
   57     
   58     Solution:
   59     reset mgmt_v3_lock_timeout to default after it be used.
   60     
   61     > fixes: 91cbcd1 glusterd-locks: misc. changes
   62     > Signed-off-by: Ren Lei <ren.lei4@zte.com.cn>
   63     
   64     Change-Id: I6a8419a7075ca2e8ce66694af31698984e2c9e4c
   65     Signed-off-by: nik-redhat <nladha@redhat.com>
   66     
   67     Co-authored-by: renlei4 <ren.lei4@zte.com.cn>
   68 
   69 commit abb42425c8b5c00d29fbe1c3632d3203f06619b6
   70 Author: schaffung <ssivakum@redhat.com>
   71 Date:   Tue Aug 10 11:26:58 2021 +0530
   72 
   73     glusterfsd: fix getcwd usage warning (#2518) (#2699)
   74     
   75     glusterfsd.c: In function ‘parse_opts’:
   76     glusterfsd.c:863:23: error: argument 1 is null but the
   77     corresponding size argument 2 value is 4096 [-Werror=nonnull]
   78       863 |                 pwd = getcwd(NULL, PATH_MAX);
   79           |                       ^~~~~~~~~~~~~~~~~~~~~~
   80     
   81     Signed-off-by: Dmitry Antipov <dantipov@cloudlinux.com>
   82     Updates: #1000
   83     Signed-off-by: srijan-sivakumar <ssivakum@redhat.com>
   84     
   85     Co-authored-by: Dmitry Antipov <dantipov@cloudlinux.com>
   86 
   87 commit d8c270d4b9c83d9b6f0b395eae06e5cbd7d4eca1
   88 Author: Ravishankar N <ravishankar@redhat.com>
   89 Date:   Mon Aug 9 13:41:15 2021 +0530
   90 
   91     afr-lock-heal-basic.t spurious failure fixes (#2452)
   92     
   93     * Fix ./tests/basic/fencing/afr-lock-heal-basic.t failure
   94     
   95     In brick-mux tests, all bricks of the volume have same pid.
   96     "generate_brick_statedump" cleans up the older statedumps
   97     with same brick pid. So successive calls to this function
   98     will delete previous brick's statedump as all bricks share
   99     same pid. So grep calls to the statedump were failing leading
  100     to failure of the .t
  101     
  102     To fix this, stored the result we need from statedump before calling
  103     next brick's statedump
  104     
  105     Fixes: #1234
  106     Change-Id: I824ed4dff79e7242b3e980364836b9af0e87a6ee
  107     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
  108     
  109     * tests: fix yet another afr-lock-heal-basic.t spurious failure (#2438)
  110     
  111     * tests: fix yet another afr-lock-heal-basic.t spurious failure
  112     
  113     From the logs, it appears as if the lock info was not present in the
  114     statedump when it was generated. Changed the logic to check for the lock
  115     info in successive statedumps within PROCESS_UP_TIMEOUT.
  116     
  117     Fixes: #2394
  118     Change-Id: I5b071299d05a8c68b02735dfd8b510b0485dc9ce
  119     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
  120     
  121     * remove sleep
  122     
  123     Change-Id: I822446222d2fbf824c6eaf42f3c72808356071e3
  124     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
  125     
  126     Co-authored-by: Pranith Kumar K <pkarampu@redhat.com>
  127 
  128 commit 8aef54b34c2e28cdf2a2fa058231708b1b322552
  129 Author: Shwetha Acharya <sacharya@redhat.com>
  130 Date:   Mon Aug 9 13:31:11 2021 +0530
  131 
  132     00-geo-rep/georep-config-upgrade.t: replace (#2698)
  133     
  134     find / -type f -name glusterfs-georep-upgrade.py
  135     with exact path of the file for better clarity
  136     and performance.
  137     
  138     Change-Id: Id4c9061312c2996076fda1ede20af2cfb2849a24
  139     Signed-off-by: Shwetha K Acharya <sacharya@redhat.com>
  140 
  141 commit 73a37d9a582458c327c9e2be54645941af3b17df
  142 Author: Shwetha Acharya <sacharya@redhat.com>
  143 Date:   Wed Jun 2 14:54:37 2021 +0530
  144 
  145     extras: upgrade script for geo-rep (#2290)
  146     
  147     The patch https://review.gluster.org/#/c/glusterfs/+/23733/( which
  148     optimizes the changelog) introduces change in dirctory structure
  149     which is above changelog files.
  150     Thus, before upgrade, old version should get updated, with respect
  151     to the corresponding changes made by the above qouted patch.
  152     
  153     This upgrade script,
  154     1) creates a temp htime file, with updated paths from the htime file.
  155        Updates temp htime file as htime file.
  156     2) places the changelog files under the required directory structure.
  157     
  158     Updates: #154
  159     Change-Id: I4b5a6cb9a9266a65972b419b329bc958de8fdf8a
  160     Signed-off-by: Shwetha K Acharya <sacharya@redhat.com>
  161 
  162 commit 92fcb9f2a2d94db64109c31483ebe22c831de550
  163 Author: Pranith Kumar K <pranith.karampuri@phonepe.com>
  164 Date:   Thu May 20 07:19:10 2021 +0530
  165 
  166     protocol/client: Initialize list head to prevent NULL de-reference
  167     
  168     fixes: #2443
  169     Change-Id: I86ef0270d41d6fb924db97fde3196d7c98c8b564
  170     Signed-off-by: Pranith Kumar K <pranith.karampuri@phonepe.com>
  171 
  172 commit a8c52aed1c58eab61b6dd24cb0dee690bde4c85a
  173 Author: Pranith Kumar Karampuri <pranith.karampuri@phonepe.com>
  174 Date:   Tue May 11 11:34:54 2021 +0530
  175 
  176     tests: Bricks should be online in shd gefore heal-full (#2416)
  177     
  178     fixes: #2404
  179     Change-Id: I2b6fa9fcff684b86936d5add25e34a649d62fea7
  180     Signed-off-by: Pranith Kumar K <pranith.karampuri@phonepe.com>
  181 
  182 commit 6028ef61d2f4078d3256a43b0efe3d2bd1a3c649
  183 Author: Tamar Shacked <tshacked@redhat.com>
  184 Date:   Sun May 9 13:20:14 2021 +0300
  185 
  186     doc: Added release 8.5 notes (#2420)
  187     
  188     * doc: Added release 8.5 notes
  189     
  190     Updates: #2340
  191     
  192     Change-Id: I890bc71a012073e119b0cef6ae07fc8a1c89ca0b
  193     Signed-off-by: Tamar Shacked <tshacked@redhat.com>
  194 
  195 commit 065a24e0a09ca44d18977a6668cfb3812c510d24
  196 Author: mohit84 <moagrawa@redhat.com>
  197 Date:   Sat May 8 07:02:33 2021 +0530
  198 
  199     core: Avoid several dict OR key is NULL message in brick logs (#1910) (#2345)
  200     
  201     Problem: dict_get_with_ref throw a message "dict or key is NULL"
  202              if dict or key is NULL.
  203     
  204     Solution: Before access a key check if dictionary is valid.
  205     
  206     > Fixes: #1909
  207     > Change-Id: I50911679142b52f854baf20c187962a2a3698f2d
  208     > Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
  209     > (Cherry picked from commit de1b26d68e31b029a59e59a47b51a7e3e6fbfe22)
  210     > (Reviewed on upstream link https://github.com/gluster/glusterfs/pull/1910)
  211     
  212     Fixes: #1909
  213     Change-Id: I50911679142b52f854baf20c187962a2a3698f2d
  214     Signed-off-by: Mohit Agrawal moagrawa@redhat.com
  215 
  216 commit 7f51663720b3aaa2251ed90e0fec1c77edd3d677
  217 Author: Pranith Kumar Karampuri <pranith.karampuri@phonepe.com>
  218 Date:   Sat May 8 05:11:39 2021 +0530
  219 
  220     protocol/client: Fix lock memory leak (#2338) (#2356)
  221     
  222     Problem-1:
  223     When an overlapping lock is issued the merged lock is not assigned the
  224     owner. When flush is issued on the fd, this particular lock is not freed
  225     leading to memory leak
  226     
  227     Fix-1:
  228     Assign the owner while merging the locks.
  229     
  230     Problem-2:
  231     On fd-destroy lock structs could be present in fdctx. For some reason
  232     with flock -x command and closing of the bash fd, it leads to this code
  233     path. Which leaks the lock structs.
  234     
  235     Fix-2:
  236     When fdctx is being destroyed in client, make sure to cleanup any lock
  237     structs.
  238     
  239     fixes: #2337
  240     Change-Id: I298124213ce5a1cf2b1f1756d5e8a9745d9c0a1c
  241     Signed-off-by: Pranith Kumar K <pranith.karampuri@phonepe.com>
  242 
  243 commit ab54656c8ebef68d0a9ee3b767a556294c201dcb
  244 Author: Pranith Kumar Karampuri <pranith.karampuri@phonepe.com>
  245 Date:   Fri May 7 15:59:23 2021 +0530
  246 
  247     mgmt/glusterd: Fix Handle leak (#2377) (#2391)
  248     
  249     fixes: #2364
  250     Change-Id: Id7d073c8fb14490e99271763c377b92df05eb64b
  251     Signed-off-by: Pranith Kumar K <pranith.karampuri@phonepe.com>
  252 
  253 commit d33c9060cea7ba41916244c12fd1a9c184dada5e
  254 Author: Nikhil Ladha <nladha@redhat.com>
  255 Date:   Thu May 6 11:03:59 2021 +0530
  256 
  257     cli: Increased spacing in cli for option table (#2341)
  258     
  259     Issue:
  260     Some options have name larger than length 40,
  261     due to which the output of command `gluster vol get <volname> all`
  262     mixes up the option, value for long option names.
  263     
  264     Fix:
  265     Increased the spacing in cli for `gluster vol get <volname> all`
  266     output to 50.
  267     
  268     Updates: #2313
  269     
  270     >Change-Id: I841730ced074547a81171a4432d15ec9c35f39cd
  271     >Signed-off-by: nik-redhat <nladha@redhat.com>
  272     
  273     >* Added separator
  274     
  275     >Change-Id: I210877c89bc468ed6a3090cd14fde7ecee1d33b6
  276     >Signed-off-by: nik-redhat <nladha@redhat.com>
  277     
  278     >* Removed separator and added space
  279     
  280     >Change-Id: Ic0eb9c9bc39a354465aabd939f72bc65be738f6c
  281     >Signed-off-by: nik-redhat <nladha@redhat.com>
  282     
  283     Change-Id: Ifc74f895b033573fb4e2698955bbc89c749fc041
  284     Signed-off-by: nik-redhat <nladha@redhat.com>
  285 
  286 commit 2759f7fda4b579bcb2c7222d8ed7877975b8a5e8
  287 Author: Pranith Kumar Karampuri <pranith.karampuri@phonepe.com>
  288 Date:   Thu May 6 10:33:16 2021 +0530
  289 
  290     features/index: Optimize link-count fetching code path (#1789) (#2387)
  291     
  292     * features/index: Optimize link-count fetching code path
  293     
  294     Problem:
  295     AFR requests 'link-count' in lookup to check if there are any pending
  296     heals. Based on this information, afr will set dirent->inode to NULL in
  297     readdirp when heals are ongoing to prevent serving bad data. When heals
  298     are completed, link-count xattr is leading to doing an opendir of
  299     xattrop directory and then reading the contents to figure out that there
  300     is no healing needed for every lookup. This was not detected until this
  301     github issue because ZFS in some cases can lead to very slow readdir()
  302     calls. Since Glusterfs does lot of lookups, this was slowing down
  303     all operations increasing load on the system.
  304     
  305     Code problem:
  306     index xlator on any xattrop operation adds index to the relevant dirs
  307     and after the xattrop operation is done, will delete/keep the index in
  308     that directory based on the value fetched in xattrop from posix. AFR
  309     sends all-zero xattrop for changelog xattrs. This is leading to
  310     priv->pending_count manipulation which sets the count back to -1. Next
  311     Lookup operation triggers opendir/readdir to find the actual link-count in
  312     lookup because in memory priv->pending_count is -ve.
  313     
  314     Fix:
  315     1) Don't add to index on all-zero xattrop for a key.
  316     2) Set pending-count to -1 when the first gfid is added into xattrop
  317        directory, so that the next lookup can compute the link-count.
  318     
  319     fixes: #1764
  320     Change-Id: I8a02c7e811a72c46d78ddb2d9d4fdc2222a444e9
  321     Signed-off-by: Pranith Kumar K <pranith.karampuri@phonepe.com>
  322     
  323     * addressed comments
  324     
  325     Change-Id: Ide42bb1c1237b525d168bf1a9b82eb1bdc3bc283
  326     Signed-off-by: Pranith Kumar K <pranith.karampuri@phonepe.com>
  327     
  328     * tests: Handle base index absence
  329     
  330     Change-Id: I3cf11a8644ccf23e01537228766f864b63c49556
  331     Signed-off-by: Pranith Kumar K <pranith.karampuri@phonepe.com>
  332     
  333     * Addressed LOCK based comments, .t comments
  334     
  335     Change-Id: I5f53e40820cade3a44259c1ac1a7f3c5f2f0f310
  336     Signed-off-by: Pranith Kumar K <pranith.karampuri@phonepe.com>
  337 
  338 commit c20e4b6da18cd5c472493bc7653ed779e94c8f42
  339 Author: mohit84 <moagrawa@redhat.com>
  340 Date:   Thu May 6 10:15:37 2021 +0530
  341 
  342     dht - fixing xattr inconsistency (#2346)
  343     
  344     The scenario of setting an xattr to a dir, killing one of the bricks,
  345     removing the xattr, bringing back the brick results in xattr
  346     inconsistency - The downed brick will still have the xattr, but the rest
  347     won't.
  348     This patch add a mechanism that will remove the extra xattrs during
  349     lookup.
  350     
  351     > fixes: #1324
  352     > Change-Id: Ifec0b7aea6cd40daa8b0319b881191cf83e031d1
  353     > Signed-off-by: Barak Sason Rofman <bsasonro@redhat.com>
  354     > Reviewed on upstream link https://review.gluster.org/#/c/glusterfs/+/24687/
  355     > (Cherry picked from commit 76017cf65433b7f42e6bfdc2eaddfc36685e2c61)
  356     
  357     fixes: #1324
  358     Change-Id: Ifec0b7aea6cd40daa8b0319b881191cf83e031d1
  359     Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
  360     
  361     Co-authored-by: Barak Sason Rofman <sason922@gmail.com>
  362 
  363 commit ddc4a9cfa3d6b24343308b85559973b850ccee09
  364 Author: Ravishankar N <ravishankar@redhat.com>
  365 Date:   Thu May 6 10:09:24 2021 +0530
  366 
  367     afr: make fsync post-op aware of inodelk count (#2273) (#2298)
  368     
  369     Problem:
  370     Since commit bd540db1e, eager-locking was enabled for fsync. But on
  371     certain VM workloads wit sharding enabled, shard xlator keeps sending
  372     fsync on the base shard. This can cause blocked inodelks from other
  373     clients (including shd) to time out due to call bail.
  374     
  375     Fix:
  376     Make afr fsync aware of inodelk count and not delay post-op + unlock
  377     when inodelk count > 1, just like writev.
  378     
  379     Code is restructured so that any fd based AFR_DATA_TRANSACTION can be made
  380     aware by setting GLUSTERFS_INODELK_DOM_COUNT in xdata request.
  381     
  382     Note: We do not know yet why VMs go in to paused state because of the
  383     blocked inodelks but this patch should be a first step in reducing the
  384     occurence.
  385     
  386     Updates: #2198
  387     Change-Id: Ib91ebdd3101d590c326e69c829cf9335003e260b
  388     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
  389 
  390 commit ce30401683c8b0dcd7826cd1521b2e28064de97c
  391 Author: Pranith Kumar Karampuri <pranith.karampuri@phonepe.com>
  392 Date:   Wed May 5 19:22:02 2021 +0530
  393 
  394     cluster/dht: Don't access local after STACK_DESTROY (#2392)
  395     
  396     There is a possibility that 'frame' could have been
  397     destroyed in dht_selfheal_dir_setattr() which can
  398     lead to local->mds_heal_fresh_lookup showing junk
  399     non-zero number.  That will lead to double
  400     STACK_DESTROY. Remembered the value of the variable
  401     before the call to fix the access.
  402     
  403     Fixes: #1214
  404     Change-Id: I37d1657798bfb549bb3887e260484d58fff42c91
  405     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
  406     
  407     Co-authored-by: Pranith Kumar K <pkarampu@redhat.com>
  408 
  409 commit 79e9b447bf82ec1ce09ed5c0c45dcc5751fa9cf2
  410 Author: Xavi Hernandez <xhernandez@users.noreply.github.com>
  411 Date:   Wed Feb 24 16:44:55 2021 +0100
  412 
  413     cluster/afr: Fix race in lockinfo (f)getxattr
  414     
  415     * cluster/afr: Fix race in lockinfo (f)getxattr
  416     
  417     A shared dictionary was updated outside the lock after having updated
  418     the number of remaining answers. This means that one thread may be
  419     processing the last answer and unwinding the request before another
  420     thread completes updating the dict.
  421     
  422         Thread 1                           Thread 2
  423     
  424         LOCK()
  425         call_cnt-- (=1)
  426         UNLOCK()
  427                                            LOCK()
  428                                            call_cnt-- (=0)
  429                                            UNLOCK()
  430                                            update_dict(dict)
  431                                            if (call_cnt == 0) {
  432                                                STACK_UNWIND(dict);
  433                                            }
  434         update_dict(dict)
  435         if (call_cnt == 0) {
  436             STACK_UNWIND(dict);
  437         }
  438     
  439     The updates from thread 1 are lost.
  440     
  441     This patch also reduces the work done inside the locked region and
  442     reduces code duplication.
  443     
  444     Fixes: #2161
  445     Change-Id: Idc0d34ab19ea6031de0641f7b05c624d90fac8fa
  446     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
  447 
  448 commit 909a9a1aa10594a7f53d2bc7fa75ca7f5d2297af
  449 Author: Xavi Hernandez <xhernandez@redhat.com>
  450 Date:   Wed Mar 17 10:59:54 2021 +0100
  451 
  452     extras: disable lookup-optimize in virt and block groups
  453     
  454     lookup-optimize doesn't provide any benefit for virtualized
  455     environments and gluster-block workloads, but it's known to cause
  456     corruption in some cases when sharding is also enabled and the volume
  457     is expanded or shrunk.
  458     
  459     For this reason, we disable lookup-optimize by default on those
  460     environments.
  461     
  462     Fixes: #2253
  463     Change-Id: I25861aa50b335556a995a9c33318dd3afb41bf71
  464     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
  465 
  466 commit 1947a59d08d103982cebb8c7caa0da63088cc970
  467 Author: Ravishankar N <ravishankar@redhat.com>
  468 Date:   Wed Mar 31 15:47:59 2021 +0530
  469 
  470     afr: remove priv->root_inode (#2244) (#2280)
  471     
  472     priv->root_inode seems to be a remenant of pump xlator and was getting
  473     populated in discover code path. thin-arbiter code used it to populate
  474     loc info but it seems that in case of some daemons like quotad, the
  475     discover path for root gfid is not hit, causing it to crash.
  476     
  477     Fix:
  478     root inode can be accessed via this->itable->root, so use that and
  479     remove priv->rot_inode instances from the afr code.
  480     
  481     Fixes: #2234
  482     Change-Id: Iec59c157f963a4dc455652a5c85a797d00cba52a
  483     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
  484 
  485 commit 8f39967e05bd3ff3fc579b197d8d147529f3b83b
  486 Author: Ravishankar N <ravishankar@redhat.com>
  487 Date:   Wed Mar 31 15:38:01 2021 +0530
  488 
  489     cli: syntax check for arbiter volume creation (#2207) (#2223)
  490     
  491     commit 8e7bfd6a58b444b26cb50fb98870e77302f3b9eb changed the syntax for
  492     arbiter volume creation to 'replica 2 arbiter 1', while still allowing
  493     the old syntax of 'replica 3 arbiter 1'. But while doing so, it also
  494     removed a conditional check, thereby allowing replica count > 3. This
  495     patch fixes it.
  496     
  497     Updates: #2192
  498     Change-Id: Ie109325adb6d78e287e658fd5f59c26ad002e2d3
  499     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
  500 
  501 commit 8330ee4ddea5e419c34596635e54a8c2dbbf2f03
  502 Author: mohit84 <moagrawa@redhat.com>
  503 Date:   Thu Mar 4 15:55:39 2021 +0530
  504 
  505     socket: Resolve ssl_ctx leak for a brick while only mgmt SSL is enabled (#2213)
  506     
  507     Problem: While only mgmt SSL is enabled for a brick process use_ssl flag
  508              is false for a brick process and socket api's cleanup ssl_ctx only
  509              while use_ssl and ssl_ctx both are valid
  510     
  511     Solution: To avoid a leak check only ssl_ctx, if it is valid cleanup
  512               ssl_ctx
  513     
  514     > Fixes: #1196
  515     > Change-Id: I2f4295478f4149dcb7d608ea78ee5104f28812c3
  516     > Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
  517     > (Reviewed on upstream link https://review.gluster.org/#/c/glusterfs/+/24366/)
  518     > (Cherry picked from commit 9873baee34afdf0c20f5fc98a7dbf2a9f07447e2)
  519     
  520     Fixes: #2210
  521     Change-Id: Icc403209cbd535c8226a23654262be63b74e351e
  522     Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
  523 
  524 commit acb67f38eeb34ca572c4258059f34a84dd303095
  525 Author: Rinku Kothiya <rkothiya@redhat.com>
  526 Date:   Wed Feb 24 15:49:57 2021 +0530
  527 
  528     doc: Added release 8.4 notes (#2185)
  529     
  530     Updates: #2183
  531     
  532     Change-Id: If15e1d6f7bbf8d76b8cdee8c6e4de18f3e818b83
  533     Signed-off-by: Rinku Kothiya <rkothiya@redhat.com>
  534 
  535 commit 2ac439ce842ce5920db5892375f5c8596baba072
  536 Author: Nikhil Ladha <nladha@redhat.com>
  537 Date:   Wed Feb 17 18:26:24 2021 +0530
  538 
  539     glusterd: Fix for shared storage in ipv6 env (#1972) (#2144)
  540     
  541     Change-Id: Ib38993724c709b35b603f9ac666630c50c932c3e
  542     Updates: #1406
  543     Signed-off-by: nik-redhat <nladha@redhat.com>
  544 
  545 commit e04627b01c97b328834602b6a6132d83f3ac96fd
  546 Author: Xavi Hernandez <xhernandez@users.noreply.github.com>
  547 Date:   Wed Feb 17 13:54:21 2021 +0100
  548 
  549     cluster/ec: Change self-heal-window-size to 4MiB by default (#2071) (#2140)
  550     
  551     The current block size used for self-heal by default is 128 KiB. This
  552     requires a significant amount of management requests for a very small
  553     portion of data healed.
  554     
  555     With this patch the block size is increased to 4 MiB. For a standard
  556     EC volume configuration of 4+2, this means that each healed block of
  557     a file will update 1 MiB on each brick.
  558     
  559     Change-Id: Ifeec4a2d54988017d038085720513c121b03445b
  560     Updates: #2067
  561     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
  562 
  563 commit d036312a7e59942d8e137b424fa2dc6012dd6ad2
  564 Author: Xavi Hernandez <xhernandez@users.noreply.github.com>
  565 Date:   Fri Feb 12 15:35:05 2021 +0100
  566 
  567     posix: fix chmod error on symlinks (#2157)
  568     
  569     After glibc 2.32, lchmod() is returning EOPNOTSUPP instead of ENOSYS when
  570     called on symlinks. The man page says that the returned code is ENOTSUP.
  571     They are the same in linux, but this patch correctly handles all errors.
  572     
  573     Fixes: #2154
  574     Change-Id: Ib3bb3d86d421cba3d7ec8d66b6beb131ef6e0925
  575     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
  576 
  577 commit 6ca27dfba90224e9dc0c025e22599c759b2d671c
  578 Author: Pranith Kumar Karampuri <pranith.karampuri@phonepe.com>
  579 Date:   Tue Feb 9 08:32:50 2021 +0530
  580 
  581     cluster/dht: Perform migrate-file with lk-owner (#2105)
  582     
  583     fixes: #1529
  584     Change-Id: I7f233ff41cda400d335f6a70ba09be736770116d
  585     Signed-off-by: Pranith Kumar K <pranith.karampuri@phonepe.com>
  586 
  587 commit 2f27e2ca1041cd8a51b62487d6f731d6ccb09d98
  588 Author: Xavi Hernandez <xhernandez@users.noreply.github.com>
  589 Date:   Sat Feb 6 01:47:08 2021 +0100
  590 
  591     dht: don't ignore xdata in fgetxattr (#2020) (#2032)
  592     
  593     DHT was passing NULL for xdata in fgetxattr() request, ignoring any
  594     data sent by upper xlators.
  595     
  596     This patch fixes the issue by sending the received xdata to lower
  597     xlators, as it's currently done for getxattr().
  598     
  599     Fixes: #1991
  600     Change-Id: If3d3f1f2ce6215f3b1acc46480e133cb4294eaec
  601     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
  602 
  603 commit f08a7de6be981b9b281047ed33016ee6475d17a1
  604 Author: Pranith Kumar Karampuri <pranith.karampuri@phonepe.com>
  605 Date:   Thu Feb 4 18:30:12 2021 +0530
  606 
  607     cluster/afr: Change default self-heal-window-size to 1MB (#2111)
  608     
  609     At the moment self-heal-window-size is 128KB. This leads to healing data
  610     in 128KB chunks. With the growth of data and the avg file sizes
  611     nowadays, 1MB seems like a better default.
  612     
  613     Change-Id: I70c42c83b16c7adb53d6b5762969e878477efb5c
  614     Fixes: #2067
  615     Signed-off-by: Pranith Kumar K <pranith.karampuri@phonepe.com>
  616 
  617 commit 628940d4da9afebf0efdc21f2d8e53d241b73eed
  618 Author: Pranith Kumar Karampuri <pranith.karampuri@phonepe.com>
  619 Date:   Thu Feb 4 18:29:12 2021 +0530
  620 
  621     cluster/dht: Allow fix-layout only on directories (#2109) (#2115)
  622     
  623     Problem:
  624     fix-layout operation assumes that the directory passed is directory i.e.
  625     layout->cnt == conf->subvolume_cnt. This will lead to a crash when
  626     fix-layout is attempted on a file.
  627     
  628     Fix:
  629     Disallow fix-layout on files
  630     
  631     fixes: #2107
  632     Change-Id: I2116b8773059f67e3260e9207e20eab3de711417
  633     Signed-off-by: Pranith Kumar K <pranith.karampuri@phonepe.com>
  634 
  635 commit e8e898fb7e385fbdaaffd9737b35396a0adb300d
  636 Author: Pranith Kumar Karampuri <pranith.karampuri@phonepe.com>
  637 Date:   Wed Feb 3 20:35:49 2021 +0530
  638 
  639     mount/fuse: Fix graph-switch when reader-thread-count is set (#1609)
  640     
  641     Problem:
  642     The current graph-switch code sets priv->handle_graph_switch to false even
  643     when graph-switch is in progress which leads to crashes in some cases
  644     
  645     Fix:
  646     priv->handle_graph_switch should be set to false only when graph-switch
  647     completes.
  648     
  649     fixes: #1539
  650     Change-Id: I5b04f7220a0a6e65c5f5afa3e28d1afe9efcdc31
  651     Signed-off-by: Pranith Kumar K <pranith.karampuri@phonepe.com>
  652 
  653 commit 3bfa520187cf37034fcd26f813385099a143a3ea
  654 Author: Ravishankar N <ravishankar@redhat.com>
  655 Date:   Tue Jan 5 20:31:24 2021 +0530
  656 
  657     dht: handle DHT_SUBVOL_STATUS_KEY in dht_pt_getxattr (#1934) (#1968)
  658     
  659     In non distribute volumes (plain replicate, ec), DHT uses pass-through
  660     FOPs (dht_pt_getxattr) instead of the usual FOPS (dht_getxattr). The
  661     pass through FOP was not handling the DHT_SUBVOL_STATUS_KEY virtual
  662     xattr because of which geo-rep session was going into a faulty state.
  663     Fixing it now.
  664     
  665     updates: #1925
  666     Change-Id: I766b5b5c047c954a9957ab78aca680eedef1ff1f
  667     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
  668 
  669 commit 3540936226df127a96d692ded2f5557adacb3f0f
  670 Author: Rinku Kothiya <rkothiya@redhat.com>
  671 Date:   Tue Dec 1 11:01:07 2020 +0530
  672 
  673     doc: Added release 8.3 notes (#1860)
  674     
  675     Updates: #1859
  676     
  677     Change-Id: I6c91be8b0ccc2c71e7edafd0c64f4d92b71efd72
  678     Signed-off-by: Rinku Kothiya <rkothiya@redhat.com>
  679 
  680 commit 04c0470319be9d36c63307d5cb0f5d2a5603272e
  681 Author: mohit84 <moagrawa@redhat.com>
  682 Date:   Sat Nov 28 18:06:07 2020 +0530
  683 
  684     posix: In the commit 2f044c4587c6db3cb82b6128f056ec2ea2bc1b98 the ret (#1858)
  685     
  686     update was missed in the function posix_get_gfid2path if
  687            GF_MALLOC is failed.
  688     
  689     Solution: Update the ret value to -1 if GF_MALLOC is failed
  690     
  691     Fixes: #1836
  692     Change-Id: I510ebf0605ee49b84ff3570948771319f283b10e
  693     Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
  694 
  695 commit e2bd0106c76299bca06385821382ec3212b63990
  696 Author: mohit84 <moagrawa@redhat.com>
  697 Date:   Fri Nov 27 17:52:03 2020 +0530
  698 
  699     posix: Use MALLOC instead of alloca to allocate memory for xattrs list (#1730) (#1827)
  700     
  701     In case of file is having huge xattrs on backend a brick process is
  702     crashed while alloca(size) limit has been crossed 256k because iot_worker
  703     stack size is 256k.
  704     
  705     Fixes: #1699
  706     Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
  707     
  708     Change-Id: I100468234f83329a7d65b43cbe4e10450c1ccecd
  709 
  710 commit e49945568ac57b6b7717d2ec46a9294dcda16472
  711 Author: Nikhil Ladha <nladha@redhat.com>
  712 Date:   Fri Nov 27 17:49:04 2020 +0530
  713 
  714     glusterd: resource leaks (#1840)
  715     
  716     * glusterd: resource leaks
  717     
  718     Issue:
  719     iobref was not freed before exiting the function
  720     if all the checks were OK, which caused the resource
  721     leak.
  722     
  723     Fix:
  724     Modified the code a bit to avoid use of an extra reference
  725     to the label, and to free the iobref and iobuf if not NULL,
  726     and then exit the function.
  727     
  728     CID: 1430118
  729     
  730     Updates: #1060
  731     
  732     Signed-off-by: nik-redhat <nladha@redhat.com>
  733     
  734     * Updated the changes
  735     
  736     Change-Id: I356fbab409ec3f22f585387819a441ded2e4ff1e
  737     Signed-off-by: nik-redhat <nladha@redhat.com>
  738 
  739 commit 00026a191e13048969a2c2736f7dea5d82e1a975
  740 Author: mohit84 <moagrawa@redhat.com>
  741 Date:   Fri Nov 27 12:26:09 2020 +0530
  742 
  743     test case ./tests/bugs/core/bug-1650403.t is getting timed out (#1847)
  744     
  745     >> The patch is backport of master patch
  746     >> https://github.com/gluster/glusterfs/commit/fc129cc2226884cbf59409256823fd9c147391be
  747     >> and
  748     >> https://github.com/mohit84/glusterfs/commit/a06b592bf5511dde4add83852441416dcda22352
  749     
  750     Fixes: #1663
  751     Change-Id: Iad9acdd1e4fe80b68c644881ad2550a194a67d2d
  752     Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
  753 
  754 commit 5aa1bb2aef4ca5f295899d16ccee30872815fc58
  755 Author: Nikhil Ladha <nladha@redhat.com>
  756 Date:   Wed Nov 25 12:25:25 2020 +0530
  757 
  758     CLI: added an error message in case of timeout (#1842)
  759     
  760     Added an error message in CLI when there are volumes present in the cluster
  761     but timeout happens while fetching them.
  762     
  763     This PR fixes #1738
  764     Backport of #1738
  765     
  766     Signed-off-by: nik-redhat <nladha@redhat.com>
  767 
  768 commit c363125154cacb97bd23800816198c6a87399e42
  769 Author: kalebskeithley <kaleb@redhat.com>
  770 Date:   Mon Nov 23 07:26:15 2020 -0500
  771 
  772     ganesha-ha: revised regex exprs for --status (#1808)
  773     
  774     better whitespace in regex
  775     
  776     This has worked for years, but somehow no longer works on rhel8
  777     
  778     Updates: #1000
  779     Change-Id: I2c1a3537573d125608334772ba1a263c55407dd4
  780     Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com>
  781 
  782 commit b3d67348f140ddf28e26754765168ff106df024a
  783 Author: mohit84 <moagrawa@redhat.com>
  784 Date:   Mon Nov 23 15:53:33 2020 +0530
  785 
  786     afr: call afr_is_lock_mode_mandatory only while xdata is valid (#1797) (#1829)
  787     
  788     afr_is_lock_mode_mandatory throws a warning message while xdata
  789     is not valid, to avoid a message call a function only while xdata
  790     is valid.
  791     
  792     Fixes: #1796
  793     Change-Id: I32d37960ea4e936ba87e65811c1792a2f1158c0d
  794     Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
  795 
  796 commit d0e5c484f7825c1582815c919929c0599d89e824
  797 Author: Ravishankar N <ravishankar@redhat.com>
  798 Date:   Mon Nov 23 12:35:47 2020 +0530
  799 
  800     glusterd: fix a bug in enabling nfs ganesha (#1813) (#1821)
  801     
  802     As detailed in the github issue,`gluster volume set Svolname ganesha.enable on`
  803     is currently broken due to a minor typo in the commit e081ac683b6a5bda548913,
  804     
  805     Fixing it now.
  806     
  807     Fixes: #1778
  808     Change-Id: I99276fedc43f40e8a439e545bd2b8d1698aa03ee
  809     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
  810     Tested-by: Strahil Nikolov <hunter86_bg@yahoo.com>
  811 
  812 commit 77eb5e8389f6912efcb5609179702a2cc5f59510
  813 Author: Ravishankar N <ravishankar@redhat.com>
  814 Date:   Tue Oct 20 11:28:31 2020 +0530
  815 
  816     fixed AFR_READ_POLICY_GFID_PID_HASH policy bug (#1561)
  817     
  818     Change-Id: Ib927a770a486c95e4b157e76ba96e9904d1a9716
  819     Fixes: #1499
  820     Signed-off-by: perrynzhou <perrynzhou@gmail.com>
  821     
  822     Co-authored-by: perrynzhou <perrynzhou@gmail.com>
  823 
  824 commit 8d7c30615a7744810a088dc4fc478b64266e4610
  825 Author: Pranith Kumar Karampuri <pranith.karampuri@phonepe.com>
  826 Date:   Tue Oct 13 15:49:00 2020 +0530
  827 
  828     rfc.sh: Move -> github flow (#1602)
  829     
  830     Change-Id: Ia4750e4f01a89b16b212d2251136da0ed738d787
  831     Fixes: #1601
  832     Signed-off-by: Pranith Kumar K <pranith.karampuri@phonepe.com>
  833 
  834 commit b8e2553b2c70da9f75d81c6f18cea41323143004
  835 Author: Kotresh HR <khiremat@redhat.com>
  836 Date:   Thu Aug 13 18:21:19 2020 +0530
  837 
  838     geo-rep: Fix string comparison
  839     
  840     Backport of:
  841       >Fixes: #1438
  842       >Change-Id: If9f1c2e89e504512bcf77608fb4a4fedb19f0399
  843       >Signed-off-by: Kotresh HR <khiremat@redhat.com>
  844     
  845     Fixes: #1438
  846     Change-Id: If9f1c2e89e504512bcf77608fb4a4fedb19f0399
  847     Signed-off-by: Shwetha K Acharya <sacharya@redhat.com>
  848     (cherry picked from commit 005d2ac1d8cc57e0e36f7cf230ae09a4f860bd16)
  849 
  850 commit 5c9b955fdc55e3cd4ba4696dbd1eb7fcb19c7a12
  851 Author: Dmitry Antipov <dmantipov@yandex.ru>
  852 Date:   Wed May 6 08:29:48 2020 +0300
  853 
  854     features/bit-rot: invalid snprintf() buffer size
  855     
  856     Found with clang-10 -Wfortify-source:
  857     
  858     bit-rot-scrub.c:1802:15: warning: 'snprintf' size argument is too large;
  859     destination buffer has size 32, but size argument is 4096 [-Wfortify-source]
  860             len = snprintf(key, PATH_MAX, "quarantine-%d", j);
  861                   ^
  862     bit-rot-scrub.c:1813:9: warning: 'snprintf' size argument is too large;
  863     destination buffer has size 32, but size argument is 4096 [-Wfortify-source]
  864             snprintf(main_key, PATH_MAX, "quarantine-%d", tmp_count);
  865     
  866     Change-Id: I9b9c09ef2223ed181d81215154345de976b82f13
  867     Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru>
  868     Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com>
  869     Fixes: #1221
  870 
  871 commit 24d35b7d031dd34ceae4816087a49038033552c8
  872 Author: Kaleb S. KEITHLEY <kkeithle@redhat.com>
  873 Date:   Tue Sep 22 08:17:05 2020 -0400
  874 
  875     build: when building with tirpc, link xlators, etc., with libtirpc
  876     
  877     uncovered on Ubuntu Groovy (20.10, ubuntu's bleeding edge devel dist),
  878     seems to now have stricter link semantics than it did when we last
  879     built 8.1 and 7.5.
  880     
  881     Many xlators actually do have direct calls to xdr_sizeof(), so strictly
  882     speaking they should be linked with libtirpc.
  883     
  884     Change-Id: Iee1fd3528fde19db397c4eae6978d9b9a2c3e17f
  885     Updates: #1002
  886     Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com>
  887 
  888 commit 895183d5a2eceabcfdd81daff87ecab1159c32c6
  889 Author: Rinku Kothiya <rkothiya@redhat.com>
  890 Date:   Wed Sep 16 07:15:41 2020 +0000
  891 
  892     doc: Added release 8.2 notes
  893     
  894     Updates: #1485
  895     
  896     Change-Id: Ia42666051df1624444ea203bf8b7c876cf78b592
  897     Signed-off-by: Rinku Kothiya <rkothiya@redhat.com>
  898 
  899 commit 85ff28ace3901a5a54d8de42d33ab2f9ac528ed8
  900 Author: Srijan Sivakumar <ssivakum@redhat.com>
  901 Date:   Tue Sep 1 12:48:48 2020 +0530
  902 
  903     Events: Fixing coverity issues.
  904     
  905     Fixing resource leak reported by coverity scan.
  906     
  907     CID: 1431237
  908     
  909     Change-Id: I2bed106b3dc4296c50d80542ee678d32c6928c25
  910     Updates: #1060
  911     Signed-off-by: Srijan Sivakumar <ssivakum@redhat.com>
  912     (cherry picked from commit ebc0253269d8a538239dd0b99d42f56ea320b0f0)
  913 
  914 commit 93d48622d9ddb96f07a8590312c2885e11751436
  915 Author: srijan-sivakumar <ssivakum@redhat.com>
  916 Date:   Sat Jul 18 05:59:09 2020 +0530
  917 
  918     Events: Socket creation after getaddrinfo and IPv4 and IPv6 packet capture
  919     
  920     Issue: Currently, the socket creation is done
  921     prior to getaddrinfo function being invoked. This
  922     can cause mismatch in the protocol and address
  923     families of the created socket and the result
  924     of the getaddrinfo api. Also, the glustereventsd
  925     UDP server by default only captures IPv4 packets
  926     hence IPv6 packets are not even captured.
  927     
  928     Code Changes:
  929     1. Modified the socket creation in such a way that
  930     the parameters taken in are dependent upon the
  931     result of the getaddrinfo function.
  932     2. Created a subclass for adding address family
  933     in glustereventsd.py for both AF_INET and AF_INET6.
  934     3. Modified addresses in the eventsapiconf.py.in
  935     
  936     Reasoning behind the approach:
  937     1. If we are using getaddrinfo function then
  938     socket creation should happen only after we
  939     check if we received back valid addresses.
  940     Hence socket creation should come after the call
  941     to getaddrinfo
  942     2. The listening server which pushes the events
  943     to the webhook has to listen for both IPv4
  944     and IPv6 messages as we would not be sure as to
  945     what address family is picked in _gf_event.
  946     
  947     Fixes: #1377
  948     Change-Id: I568dcd1a977c8832f0fef981e1f81cac7043c760
  949     Signed-off-by: srijan-sivakumar <ssivakum@redhat.com>
  950     (cherry picked from commit 7c309928591deb8d0188793677958226ac03897a)
  951 
  952 commit b4cc0988d5e9a5bf354dd4c2cb254ce52546facb
  953 Author: nik-redhat <nladha@redhat.com>
  954 Date:   Thu Sep 10 14:55:35 2020 +0530
  955 
  956     glusterd: readdir-ahead off by default
  957     
  958     Changing the default value of readdir-ahead to
  959     off, but it can be enabled/disabled later on if with
  960     gluster vol set <volname> performance.readdir-ahead enabel/disable
  961     command.
  962     
  963     Updates: #1472
  964     
  965     Change-Id: Idb3e16e8be98d7a811fc8e5d09906919ef50fbab
  966     Signed-off-by: nik-redhat <nladha@redhat.com>
  967     (cherry picked from commit 84a4cf76219b6187fc625740d1a1ebbe40e9f22c)
  968 
  969 commit 68db6b60f621d371c4059a7ee728ebb267854708
  970 Author: nik-redhat <nladha@redhat.com>
  971 Date:   Wed Aug 26 15:08:56 2020 +0530
  972 
  973     glusterd: cksum mismatch on upgrading to latest gluster
  974     
  975     Issue:
  976     In gluster versions less than 7, the checksums were calculated
  977     whether or not the quota is enabled or not, and that cksum value
  978     was also getting stored in the quota.cksum file. But, from gluster
  979     7 version onwards cksum was calculated only if the quota is enabled.
  980     Due to this, the cksums in quota.cksum files differ after upgrading.
  981     
  982     Fix:
  983     Added a check to see if the OP_VERSION is less than 7 then, follow
  984     the previous method otherwise, move as per the latest changes for
  985     cksum calculation.
  986     
  987     This changes for the cksum calculation was done in
  988     this commit : https://github.com/gluster/glusterfs/commit/3b5eb592f5
  989     
  990     Updates: #1332
  991     
  992     Change-Id: I7a95e5e5f4d4be4983fb7816225bf9187856c003
  993     Signed-off-by: nik-redhat <nladha@redhat.com>
  994     (cherry picked from commit 865cca1190e233381f975ff36118f46e29477dcf)
  995     Signed-off-by: nik-redhat <nladha@redhat.com>
  996 
  997 commit a5d9edce9b59ee00d2a4027fafba126e82e2fcfd
  998 Author: Xavi Hernandez <xhernandez@redhat.com>
  999 Date:   Fri Sep 4 14:49:50 2020 +0200
 1000 
 1001     open-behind: implement create fop
 1002     
 1003     Open behind didn't implement create fop. This caused that files created
 1004     were not accounted for the number of open fd's. This could cause future
 1005     opens to be delayed when they shouldn't.
 1006     
 1007     This patch implements the create fop. It also fixes a problem when
 1008     destroying the stack: when frame->local was not NULL, STACK_DESTROY()
 1009     tried to mem_put() it, which is not correct.
 1010     
 1011     Fixes: #1440
 1012     Change-Id: Ic982bad07d4af30b915d7eb1fbcef7a847a45869
 1013     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
 1014 
 1015 commit 473453c4e2b1b6fc94edbce438dd9a3c0ea58c67
 1016 Author: Amar Tumballi <amar@kadalu.io>
 1017 Date:   Tue Aug 18 14:08:20 2020 +0530
 1018 
 1019     tests: provide an option to mark tests as 'flaky'
 1020     
 1021     * also add some time gap in other tests to see if we get things properly
 1022     * create a directory 'tests/000/', which can host any tests, which are flaky.
 1023     * move all the tests mentioned in the issue to above directory.
 1024     * as the above dir gets tested first, all flaky tests would be reported quickly.
 1025     * change `run-tests.sh` to continue tests even if flaky tests fail.
 1026     
 1027     Reference: gluster/project-infrastructure#72
 1028     Updates: #1000
 1029     Change-Id: Ifdafa38d083ebd80f7ae3cbbc9aa3b68b6d21d0e
 1030     Signed-off-by: Amar Tumballi <amar@kadalu.io>
 1031     (cherry picked from 097db13c11390174c5b9f11aa0fd87eca1735871)
 1032 
 1033 commit 635dcf82505efcdeaf01c4e0450a157b533099ba
 1034 Author: Ravishankar N <ravishankar@redhat.com>
 1035 Date:   Tue Sep 1 11:36:42 2020 +0530
 1036 
 1037     libglusterfs: fix dict leak
 1038     
 1039     Problem:
 1040     gf_rev_dns_lookup_cached() allocated struct dnscache->dict if it was null
 1041     but the freeing was left to the caller.
 1042     
 1043     Fix:
 1044     Moved dict allocation and freeing into corresponding init and fini
 1045     routines so that its easier for the caller to avoid such leaks.
 1046     
 1047     Updates: #1000
 1048     Change-Id: I90d6a6f85ca2dd4fe0ab461177aaa9ac9c1fbcf9
 1049     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
 1050     (cherry picked from commit 079f7a7d8a2bd85070c1da4dde2452ca82a1cdbb)
 1051 
 1052 commit f9b8462ba212e0fd572efdf6ade03f4d5c53d11e
 1053 Author: Rinku Kothiya <rkothiya@redhat.com>
 1054 Date:   Tue Aug 25 12:31:20 2020 +0000
 1055 
 1056     doc: Updated release 8.1 notes
 1057     
 1058     Updates: #1318
 1059     
 1060     Change-Id: I87787a1aaf59302ad045ed6d2562920e17654678
 1061     Signed-off-by: Rinku Kothiya <rkothiya@redhat.com>
 1062 
 1063 commit ab40a26dcd9ce8b676f482bf751e57024e227e89
 1064 Author: Rinku Kothiya <rkothiya@redhat.com>
 1065 Date:   Sat Aug 22 17:23:25 2020 +0000
 1066 
 1067     doc: Added release 8.1 notes
 1068     
 1069     Updates: #1318
 1070     
 1071     Change-Id: I14d589bd9af85bdd4ae02902e41d4c5f7d930358
 1072     Signed-off-by: Rinku Kothiya <rkothiya@redhat.com>
 1073 
 1074 commit f120311737cf681c36423d8551e2e218509cd5f7
 1075 Author: Ravishankar N <ravishankar@redhat.com>
 1076 Date:   Wed Aug 19 11:14:25 2020 +0530
 1077 
 1078     afr: add null check for thin-arbiter gfid.
 1079     
 1080     Problem:
 1081     Lookup/creation of thin-arbiter ID file happens in background during
 1082     mounting. On new volumes, if the  ID file creation is in progress, and a
 1083     FOP fails on data brick, a post-op (xattrop) is attemtped on TA. Since
 1084     the TA file's gfid is null at this point, the ASSERT checks in protocol/
 1085     client causes a crash.
 1086     
 1087     Fix:
 1088     Given that we decided to do Lookup/creation of thin-arbiter in
 1089     background, fail the other AFR FOPS on TA if the ID file's gfid is null
 1090     instead of winding it down to protocol/client.
 1091     
 1092     Also remove afr_changelog_thin_arbiter_post_op() which seems to be dead
 1093     code.
 1094     
 1095     Updates: #763
 1096     Change-Id: I70dc666faf55cc5c8f7cf8e7d36085e4fa399c4d
 1097     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
 1098     (cherry picked from commit f9b5074394e3d2f3b6728aab97230ba620879426)
 1099 
 1100 commit 4398db9d70f34e9a8af88fe3de564a906db7b182
 1101 Author: Xavi Hernandez <xhernandez@redhat.com>
 1102 Date:   Wed Aug 19 23:27:38 2020 +0200
 1103 
 1104     open-behind: fix call_frame leak
 1105     
 1106     When an open was delayed, a copy of the frame was created because the
 1107     current frame was used to unwind the "fake" open. When the open was
 1108     actually sent, the frame was correctly destroyed. However if the file
 1109     was closed before needing to send the open, the frame was not destroyed.
 1110     
 1111     This patch correctly destroys the frame in all cases.
 1112     
 1113     Change-Id: I8c00fc7f15545c240e8151305d9e4cf06d653926
 1114     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
 1115     Fixes: #1440
 1116 
 1117 commit e173c5b0ee32c210a7d36f03f1847c42218a62e5
 1118 Author: Mohit Agrawal <moagrawa@redhat.com>
 1119 Date:   Mon Jul 27 18:08:00 2020 +0530
 1120 
 1121     posix: Implement a janitor thread to close fd
 1122     
 1123     Problem: In the commit fb20713b380e1df8d7f9e9df96563be2f9144fd6 we use
 1124              syntask to close fd but we have found the patch is reducing the
 1125              performance
 1126     
 1127     Solution: Use janitor thread to close fd's and save the pfd ctx into
 1128               ctx janitor list and also save the posix_xlator into pfd object to
 1129               avoid the race condition during cleanup in brick_mux environment
 1130     
 1131     Change-Id: Ifb3d18a854b267333a3a9e39845bfefb83fbc092
 1132     Fixes: #1396
 1133     Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
 1134     (cherry picked from commit 41b9616435cbdf671805856e487e373060c9455b)
 1135 
 1136 commit 05060c9664153beb392206ae05a498d4d4178f5f
 1137 Author: Leonid Ishimnikov <lishim@fastmail.com>
 1138 Date:   Thu Aug 13 15:37:50 2020 -0400
 1139 
 1140     glusterd: dump SSL error stack on disconnect
 1141     
 1142     Problem:  When a non-SSL connection is attempted on an SSL-enabled
 1143               management port, unrelated peers are subsequently disconnected
 1144               from the node with a misleading error message.
 1145     Cause:    A non-SSL client causes OpenSSL to push a wrong version error
 1146               into its thread-local error stack, but this error is never
 1147               cleared, and it lingers in the stack until the thread is used
 1148               by another SSL session, and a certain condition requires the error
 1149               stack to be examined, at which time the old error is discovered and
 1150               the connection is terminated.
 1151     Solution: Log and clear the error stack upon terminating the connection.
 1152     
 1153     Change-Id: I82f3a723285df24dafc88850ae4fca65b69f6ae4
 1154     Fixes: #1418
 1155     Signed-off-by: Leonid Ishimnikov <lishim@fastmail.com>
 1156     (cherry picked from commit bb5801d1480314e09b4203d2525bd01aada5c683)
 1157 
 1158 commit c5fc58c8cb01753e2fed173c76aea1e9cc333862
 1159 Author: Vinayakswami Hariharmath <vharihar@redhat.com>
 1160 Date:   Thu Aug 6 14:39:59 2020 +0530
 1161 
 1162     features/shard: optimization over shard lookup in case of prealloc
 1163     
 1164     Assume that we are preallocating a VM of size 1TB with a shard
 1165     block size of 64MB then there will be ~16k shards.
 1166     
 1167     This creation happens in 2 steps shard_fallocate() path i.e
 1168     
 1169     1. lookup for the shards if any already present and
 1170     2. mknod over those shards do not exist.
 1171     
 1172     But in case of fresh creation, we dont have to lookup for all
 1173     shards which are not present as the the file size will be 0.
 1174     Through this, we can save lookup on all shards which are not
 1175     present. This optimization is quite useful in the case of
 1176     preallocating big vm.
 1177     
 1178     Also if the file is already present and the call is to
 1179     extend it to bigger size then we need not to lookup for non-
 1180     existent shards. Just lookup preexisting shards, populate
 1181     the inodes and issue mknod on extended size.
 1182     
 1183     Fixes: #1425
 1184     Change-Id: I60036fe8302c696e0ca80ff11ab0ef5bcdbd7880
 1185     Signed-off-by: Vinayakswami Hariharmath <vharihar@redhat.com>
 1186     (cherry picked from commit 2ede911d07c6dc07a0f729526ab590ace77341ae)
 1187 
 1188 commit 8ef4b79162a0409023b10a15561c84606e0b3ae0
 1189 Author: Krutika Dhananjay <kdhananj@redhat.com>
 1190 Date:   Mon May 4 14:30:57 2020 +0530
 1191 
 1192     extras: Modify group 'virt' to include network-related options
 1193     
 1194     This is needed to work around an issue seen where vms running on
 1195     online hosts are getting killed when a different host is rebooted
 1196     in ovirt-gluster hyperconverged environments. Actual RCA is quite
 1197     lengthy and documented in the github issue. Please refer to it
 1198     for more details.
 1199     
 1200     Change-Id: Ic25b5f50144ad42458e5c847e1e7e191032396c1
 1201     Fixes: #1217
 1202     Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
 1203     (cherry picked from commit 5391f16fc4aa00f75af2a4c2707768370ace5f6c)
 1204 
 1205 commit 7b372cdeed876e68293620c25c6821324068fb54
 1206 Author: Ashish Pandey <aspandey@redhat.com>
 1207 Date:   Thu Jul 23 11:07:32 2020 +0530
 1208 
 1209     cluster/ec: Remove stale entries from indices/xattrop folder
 1210     
 1211     Problem:
 1212     If a gfid is present in indices/xattrop folder while
 1213     the file/dir is actaully healthy and all the xattrs are healthy,
 1214     it causes lot of lookups by shd on an entry which does not need
 1215     to be healed.
 1216     This whole process eats up lot of CPU usage without doing meaningful
 1217     work.
 1218     
 1219     Solution:
 1220     Set trusted.ec.dirty xattr of the entry so that actual heal process
 1221     happens and at the end of it, during unset of dirty, gfid enrty from
 1222     indices/xattrop will be removed.
 1223     
 1224     Change-Id: Ib1b9377d8dda384bba49523e9ff6ba9f0699cc1b
 1225     Fixes: #1385
 1226     Signed-off-by: Ashish Pandey <aspandey@redhat.com>
 1227     (cherry picked from commit ba1b0a471dec968633f89c7f790b099fb4ad700d)
 1228 
 1229 commit 7ff51badda5cbcbaa17f729d1e4ab715c462396a
 1230 Author: Mohit Agrawal <moagrawa@redhat.com>
 1231 Date:   Sat Aug 1 09:28:47 2020 +0530
 1232 
 1233     glusterd: Increase buffer length to save multiple hostnames in peer file
 1234     
 1235     Problem: At the time of handling friend update request glusterd updates peer
 1236              file and if DNS has returned multiple hostnames for the same IP, glusterd
 1237              saves all hostnames in peer file.In commit 1fa089e7a2b180e0bdcc1e7e09a63934a2a0c0ef
 1238              We changed the approach to save all key value pairs in single shot.
 1239              In case of a buffer is not having space to store the hostnames glusterd
 1240              writes partial hostname in peer file.
 1241     
 1242     Solution: To avoid the failure increase the buffer length
 1243     
 1244     Change-Id: Iee969d165333e9c5ba69431d474c541b8f12d442
 1245     Fixes: #1407
 1246     Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
 1247     (cherry picked from commit 6e8e73a06d71382f8f6e3cd83fe72692d19e66ba)
 1248 
 1249 commit 4be26c88732f55b38da171c86334eddbdaac5c14
 1250 Author: Sunny Kumar <sunkumar@redhat.com>
 1251 Date:   Tue May 19 16:13:01 2020 +0100
 1252 
 1253     geo-rep: Fix corner case in rename on mkdir during hybrid crawl
 1254     
 1255     Problem:
 1256     The issue is being hit during hybrid mode while handling rename on slave.
 1257     In this special case the rename is recorded as mkdir and geo-rep process it
 1258     by resolving the path form backend.
 1259     
 1260     While resolving the backend path during this special handling one corner case is not considered.
 1261     
 1262     <snip>
 1263     Traceback (most recent call last):
 1264       File "/usr/libexec/glusterfs/python/syncdaemon/repce.py", line 118, in worker
 1265         res = getattr(self.obj, rmeth)(*in_data[2:])
 1266       File "/usr/libexec/glusterfs/python/syncdaemon/resource.py", line 588, in entry_ops
 1267         src_entry = get_slv_dir_path(slv_host, slv_volume, gfid)
 1268       File "/usr/libexec/glusterfs/python/syncdaemon/syncdutils.py", line 710, in get_slv_dir_path
 1269         dir_entry = os.path.join(pfx, pargfid, basename)
 1270       File "/usr/lib64/python2.7/posixpath.py", line 75, in join
 1271         if b.startswith('/'):
 1272     AttributeError: 'int' object has no attribute 'startswith'
 1273     
 1274     In pyhthon3:
 1275     Traceback (most recent call last):
 1276       File "<stdin>", line 1, in <module>
 1277       File "/usr/lib64/python3.8/posixpath.py", line 90, in join
 1278         genericpath._check_arg_types('join', a, *p)
 1279       File "/usr/lib64/python3.8/genericpath.py", line 152, in _check_arg_types
 1280         raise TypeError(f'{funcname}() argument must be str, bytes, or '
 1281     TypeError: join() argument must be str, bytes, or os.PathLike object, not 'int'
 1282     </snip>
 1283     
 1284     Backport of:
 1285       >Ptach link: https://review.gluster.org/#/c/glusterfs/+/24468/
 1286       >Change-Id: I8b926899c60ad8c4ffc886d57028ba70fd21e332
 1287       >Fixes: #1250
 1288       >Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
 1289     
 1290     Change-Id: I8b926899c60ad8c4ffc886d57028ba70fd21e332
 1291     Fixes: #1250
 1292     Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
 1293     
 1294     (cherry picked from commit 27f5c8ba844e9da54fc1304df4ffe015a3bbb9bd)
 1295     
 1296     Change-Id: I171eb9ad4e30f49cfe86cb258918682d3c0f5af9
 1297 
 1298 commit 269ece312c9fd890c74c46e79de70efe1720752c
 1299 Author: Xavi Hernandez <xhernandez@redhat.com>
 1300 Date:   Thu Jul 2 18:08:52 2020 +0200
 1301 
 1302     cluster/ec: Improve detection of new heals
 1303     
 1304     When EC successfully healed a directory it assumed that maybe other
 1305     entries inside that directory could have been created, which could
 1306     require additional heal cycles. For this reason, when the heal happened
 1307     as part of one index heal iteration, it triggered a new iteration.
 1308     
 1309     The problem happened when the directory was healthy, so no new entries
 1310     were added, but its index entry was not removed for some reason. In
 1311     this case self-heal started and endless loop healing the same directory
 1312     continuously, cause high CPU utilization.
 1313     
 1314     This patch improves detection of new files added to the heal index so
 1315     that a new index heal iteration is only triggered if there is new work
 1316     to do.
 1317     
 1318     Change-Id: I2355742b85fbfa6de758bccc5d2e1a283c82b53f
 1319     Fixes: #1354
 1320     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
 1321 
 1322 commit 19dc7b37fc9f6d6037958e5dd3c0c6a4a993e2af
 1323 Author: Krutika Dhananjay <kdhananj@redhat.com>
 1324 Date:   Thu Sep 12 11:07:10 2019 +0530
 1325 
 1326     features/shard: Convert shard block indices to uint64
 1327     
 1328     This patch fixes a crash in FOPs that operate on really large sharded
 1329     files where number of participant shards could sometimes exceed
 1330     signed int32 max.
 1331     
 1332     The patch also adds GF_ASSERTs to ensure that number of participating
 1333     shards is always greater than 0 for files that do have more than one
 1334     shard.
 1335     
 1336     Change-Id: I354de58796f350eb1aa42fcdf8092ca2e69ccbb6
 1337     Fixes: #1348
 1338     Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
 1339     (cherry picked from commit cdf01cc47eb2efb427b5855732d9607eec2abc8a)
 1340 
 1341 commit 96e7ff7396a8e18ca69d5198be0a8d29bcc37129
 1342 Author: Vinayakswami Hariharmath <vharihar@redhat.com>
 1343 Date:   Wed Jun 3 18:58:56 2020 +0530
 1344 
 1345     features/shard: Use fd lookup post file open
 1346     
 1347     Issue:
 1348     When a process has the open fd and the same file is
 1349     unlinked in middle of the operations, then file based
 1350     lookup fails with ENOENT or stale file
 1351     
 1352     Solution:
 1353     When the file already open and fd is available, use fstat
 1354     to get the file attributes
 1355     
 1356     Change-Id: I0e83aee9f11b616dcfe13769ebfcda6742e4e0f4
 1357     Fixes: #1281
 1358     Signed-off-by: Vinayakswami Hariharmath <vharihar@redhat.com>
 1359     (cherry picked from commit 71dd19f710b81136f318b3a95ae430971198ee70)
 1360 
 1361 commit 6b10b33f8a9bce054fec980583cc597f5a438bb5
 1362 Author: Soumya Koduri <skoduri@redhat.com>
 1363 Date:   Thu Jul 2 02:07:56 2020 +0530
 1364 
 1365     Issue with gf_fill_iatt_for_dirent
 1366     
 1367     In "gf_fill_iatt_for_dirent()", while calculating inode_path for loc,
 1368     the inode should be of parent's. Instead it is loc.inode which results in error
 1369      and eventually lookup/readdirp fails.
 1370     
 1371     This patch fixes the same.
 1372     
 1373     Change-Id: Ied086234a4634e8cb13520521ac547c87b3c76b5
 1374     Fixes: #1351
 1375     Signed-off-by: Soumya Koduri <skoduri@redhat.com>
 1376     (cherry picked from commit ab8308333aaf033e07dbbdf2f69f9313a7e311f3)
 1377 
 1378 commit e8aedcd40f9f24e5b821e1539275e40ebfccca94
 1379 Author: Pranith Kumar K <pkarampu@redhat.com>
 1380 Date:   Fri May 29 14:24:53 2020 +0530
 1381 
 1382     cluster/afr: Delay post-op for fsync
 1383     
 1384     Problem:
 1385     AFR doesn't delay post-op for fsync fop. For fsync heavy workloads
 1386     this leads to un-necessary fxattrop/finodelk for every fsync leading
 1387     to bad performance.
 1388     
 1389     Fix:
 1390     Have delayed post-op for fsync. Add special flag in xdata to indicate
 1391     that afr shouldn't delay post-op in cases where either the
 1392     process will terminate or graph-switch would happen. Otherwise it leads
 1393     to un-necessary heals when the graph-switch/process-termination
 1394     happens before delayed-post-op completes.
 1395     
 1396     Fixes: #1253
 1397     Change-Id: I531940d13269a111c49e0510d49514dc169f4577
 1398     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 1399 
 1400 commit 30b95ff9cdec72d9089f4882dafca447ae3174f1
 1401 Author: Kaleb S. KEITHLEY <kkeithle@redhat.com>
 1402 Date:   Thu Jul 2 15:52:15 2020 -0400
 1403 
 1404     api: libgfapi symbol versions break LTO in Fedora rawhide/f33
 1405     
 1406     The way symbol versions are implemented is incompatible with gcc-10 and LTO.
 1407     
 1408     Fedora provenpackager Jeff Law (law [at] redhat.com) writes in the
 1409     Fedora dist-git glusterfs.spec:
 1410       This package uses top level ASM constructs which are incompatible with LTO.
 1411       Top level ASMs are often used to implement symbol versioning. gcc-10
 1412       introduces a new mechanism for symbol versioning which works with LTO.
 1413       Converting packages to use that mechanism instead of toplevel ASMs is
 1414       recommended.
 1415     
 1416     In particular, note that the version of gluster in Fedora rawhide/f33 is
 1417     glusterfs-8.0RC0. Once this fix is merged it will be necessary to backport
 1418     it to the release-8 branch.
 1419     
 1420     At the time that gfapi symbol versions were first implemented we copied
 1421     the GNU libc (glibc) symbol version implementation following Uli Drepper's
 1422     symbol versioning HOWTO.
 1423     
 1424     Now gcc-10 has a symver attribute that can be used instead. (Maybe it
 1425     has been there all along?)
 1426     
 1427     Both the original implemenation and this implemenation yield the same
 1428     symbol versions. This can be seen by running
 1429       `nm -D --with-symbol-versions libgfapi.so`
 1430     on the libgfapi.so built before and after applying this fix.
 1431     
 1432     Change-Id: I05fda580afacfff1bfc07be810dd1afc08a92fb8
 1433     Fixes: #1352
 1434     Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com>
 1435 
 1436 commit 2e1e4168ab6b6b3ebe9e3dfb227fb7631e5c1aa4
 1437 Author: Rinku Kothiya <rkothiya@redhat.com>
 1438 Date:   Sun Jul 5 03:17:04 2020 +0000
 1439 
 1440     doc: Updated release 8.0 notes
 1441     
 1442     Updates: #1180
 1443     
 1444     Change-Id: I6e5c85f2714896704949d9a99b06eefdde15d633
 1445     Signed-off-by: Rinku Kothiya <rkothiya@redhat.com>
 1446 
 1447 commit 49e3a380055f955303ed6d847aedf1ef61ef71d5
 1448 Author: Krutika Dhananjay <kdhananj@redhat.com>
 1449 Date:   Fri May 22 13:25:26 2020 +0530
 1450 
 1451     features/shard: Aggregate file size, block-count before unwinding removexattr
 1452     
 1453     Posix translator returns pre and postbufs in the dict in {F}REMOVEXATTR fops.
 1454     These iatts are further cached at layers like md-cache.
 1455     Shard translator, in its current state, simply returns these values without
 1456     updating the aggregated file size and block-count.
 1457     
 1458     This patch fixes this problem.
 1459     
 1460     Change-Id: I4b2dd41ede472c5829af80a67401ec5a6376d872
 1461     Fixes: #1243
 1462     Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
 1463     (cherry picked from commit 32519525108a2ac6bcc64ad931dc8048d33d64de)
 1464 
 1465 commit d79e85e4ffa52c40babec777578f98706e334209
 1466 Author: Kaleb S. KEITHLEY <kkeithle@redhat.com>
 1467 Date:   Thu Jun 11 06:46:56 2020 -0400
 1468 
 1469     packaging: refactor to align with common practices
 1470     
 1471     Apparently some sdditional Obsoletes: are required
 1472     
 1473     Change-Id: I919ae5a0fcc6f720e3eab4784af36977b9eef044
 1474     Fixes: #1126
 1475     Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com>
 1476 
 1477 commit 0a97167f850c2fa5ac8f9952f35933fddada3140
 1478 Author: Kaleb S. KEITHLEY <kkeithle@redhat.com>
 1479 Date:   Thu May 14 06:47:17 2020 -0400
 1480 
 1481     packaging: refactor to align with common practices
 1482     
 1483     The claim that Fedora package guidelines do not require this
 1484     scheme is a non-argument. Not only do they not require it, they
 1485     don't prohibit it either. (And you can't prove a negative. It's
 1486     a specious argument.)
 1487     
 1488     Change-Id: I7748c7531d52dedd71b3a7f5df049742258a6aba
 1489     Fixes: #1126
 1490     Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com>
 1491 
 1492 commit 80e088be4470d86e5c63750f611a39e487a591b3
 1493 Author: Xavi Hernandez <xhernandez@redhat.com>
 1494 Date:   Tue May 12 23:54:54 2020 +0200
 1495 
 1496     open-behind: rewrite of internal logic
 1497     
 1498     There was a critical flaw in the previous implementation of open-behind.
 1499     
 1500     When an open is done in the background, it's necessary to take a
 1501     reference on the fd_t object because once we "fake" the open answer,
 1502     the fd could be destroyed. However as long as there's a reference,
 1503     the release function won't be called. So, if the application closes
 1504     the file descriptor without having actually opened it, there will
 1505     always remain at least 1 reference, causing a leak.
 1506     
 1507     To avoid this problem, the previous implementation didn't take a
 1508     reference on the fd_t, so there were races where the fd could be
 1509     destroyed while it was still in use.
 1510     
 1511     To fix this, I've implemented a new xlator cbk that gets called from
 1512     fuse when the application closes a file descriptor.
 1513     
 1514     The whole logic of handling background opens have been simplified and
 1515     it's more efficient now. Only if the fop needs to be delayed until an
 1516     open completes, a stub is created. Otherwise no memory allocations are
 1517     needed.
 1518     
 1519     Correctly handling the close request while the open is still pending
 1520     has added a bit of complexity, but overall normal operation is simpler.
 1521     
 1522     Change-Id: I6376a5491368e0e1c283cc452849032636261592
 1523     Fixes: #1225
 1524     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
 1525 
 1526 commit 6872b75caf7a7bc945dba73c381236f344548728
 1527 Author: Krutika Dhananjay <kdhananj@redhat.com>
 1528 Date:   Fri May 15 11:29:36 2020 +0530
 1529 
 1530     features/shard: Aggregate size, block-count in iatt before unwinding setxattr
 1531     
 1532     Posix translator returns pre and postbufs in the dict in {F}SETXATTR fops.
 1533     These iatts are further cached at layers like md-cache.
 1534     Shard translator, in its current state, simply returns these values without
 1535     updating the aggregated file size and block-count.
 1536     
 1537     This patch fixes this problem.
 1538     
 1539     Change-Id: I4da0eceb4235b91546df79270bcc0af8cd64e9ea
 1540     Fixes: #1243
 1541     Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
 1542     (cherry picked from commit 29ec66c6ab77e2d6893c6e213a3d1fb148702c99)
 1543 
 1544 commit 215cf1728e029d062b21b712b6c5356b3bf73794
 1545 Author: Ravishankar N <ravishankar@redhat.com>
 1546 Date:   Wed May 27 20:50:46 2020 +0530
 1547 
 1548     afr: more quorum checks in lookup and new entry marking
 1549     
 1550     Problem: See github issue for details.
 1551     
 1552     Fix:
 1553     -In lookup if the entry exists in 2 out of 3 bricks, don't fail the
 1554     lookup with ENOENT just because there is an entrylk on the parent.
 1555     Consider quorum before deciding.
 1556     
 1557     -If entry FOP does not succeed on quorum no. of bricks, do not perform
 1558     new entry mark.
 1559     
 1560     Fixes: #1303
 1561     Change-Id: I56df8c89ad53b29fa450c7930a7b7ccec9f4a6c5
 1562     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
 1563     (cherry picked from commit c4a6748f25d2c1ab3ebcf89952278ebf94c8d371)
 1564 
 1565 commit 8c1008646ec7e62b33d92184358eb455f7e44350
 1566 Author: Xavi Hernandez <xhernandez@redhat.com>
 1567 Date:   Fri Jun 12 00:06:36 2020 +0200
 1568 
 1569     locks: prevent deletion of locked entries
 1570     
 1571     To keep consistency inside transactions started by locking an entry or
 1572     an inode, this change delays the removal of entries that are currently
 1573     locked by one or more clients. Once all locks are released, the removal
 1574     is processed.
 1575     
 1576     It has also been improved the detection of stale inodes in the locking
 1577     code of EC.
 1578     
 1579     Fixes: #990
 1580     Change-Id: Ic8ba23d9480f80c7f74e7a310bf8a15922320fd5
 1581     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
 1582 
 1583 commit 3e8c565504aa9f61f0e362072170e8d8e974c2c7
 1584 Author: karthik-us <ksubrahm@redhat.com>
 1585 Date:   Thu May 21 15:18:59 2020 +0530
 1586 
 1587     cluster/afr: Prioritize ENOSPC over other errors
 1588     
 1589     Problem:
 1590     In a replicate/arbiter volume if file creations or writes fails on
 1591     quorum number of bricks and on one brick it is due to ENOSPC and
 1592     on other brick it fails for a different reason, it may fail with
 1593     errors other than ENOSPC in some cases.
 1594     
 1595     Fix:
 1596     Prioritize ENOSPC over other lesser priority errors and do not set
 1597     op_errno in posix_gfid_set if op_ret is 0 to avoid receiving any
 1598     error_no which can be misinterpreted by __afr_dir_write_finalize().
 1599     
 1600     Also removing the function afr_has_arbiter_fop_cbk_quorum() which
 1601     might consider a successful reply form a single brick as quorum
 1602     success in some cases, whereas we always need fop to be successful
 1603     on quorum number of bricks in arbiter configuration.
 1604     
 1605     Change-Id: I106e267f8b9451f681022f1cccb410d9bc824c08
 1606     Fixes: #1254
 1607     Signed-off-by: karthik-us <ksubrahm@redhat.com>
 1608     (cherry picked from commit fa63b45ca5edf172b1b89b28b5db3c5129cc57b6)
 1609 
 1610 commit 18bd1bdaa6ea5d589b21865769d6183e4e201006
 1611 Author: Xavi Hernandez <xhernandez@redhat.com>
 1612 Date:   Fri Apr 17 12:10:05 2020 +0200
 1613 
 1614     md-cache: fix several NULL dereferences
 1615     
 1616     This patch includes the following CID from Coverity Scan:
 1617     
 1618       * 1425196
 1619       * 1425197
 1620       * 1425198
 1621       * 1425199
 1622       * 1525200
 1623     
 1624     Change-Id: Iddcfea449d3dd56d4dfcc39f4c3c608518e611e4
 1625     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
 1626     Updates: #1060
 1627     (cherry picked from commit b53ba17dbfd2d18c10e2c308b8899d36726ab440)
 1628 
 1629 commit 9399faa68968bc2aaff7390580cb072f2926c8ed
 1630 Author: Rinku Kothiya <rkothiya@redhat.com>
 1631 Date:   Thu Apr 23 14:32:02 2020 +0000
 1632 
 1633     doc: Added release 8.0 notes
 1634     
 1635     Updates: #1180
 1636     
 1637     Change-Id: If3ff097e299261f2b647d5faa85d268b0a86908c
 1638     Signed-off-by: Rinku Kothiya <rkothiya@redhat.com>
 1639 
 1640 commit 67cd2b1ab5c9a8db637506b9a6be57fc79672c99
 1641 Author: Xavi Hernandez <xhernandez@redhat.com>
 1642 Date:   Thu Apr 30 11:19:01 2020 +0200
 1643 
 1644     syncop: improve scaling and implement more tools
 1645     
 1646     The current scaling of the syncop thread pool is not working properly
 1647     and can leave some tasks in the run queue more time than necessary
 1648     when the maximum number of threads is not reached.
 1649     
 1650     This patch provides a better scaling condition to react faster to
 1651     pending work.
 1652     
 1653     Condition variables and sleep in the context of a synctask have also
 1654     been implemented. Their purpose is to replace regular condition
 1655     variables and sleeps that block synctask threads and prevent other
 1656     tasks to be executed.
 1657     
 1658     The new features have been applied to several places in glusterd.
 1659     
 1660     Change-Id: Ic50b7c73c104f9e41f08101a357d30b95efccfbf
 1661     Fixes: #1116
 1662     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
 1663 
 1664 commit 49713a7ea1d30613a8810b85eca8ab35f6e9fa8a
 1665 Author: Kaleb S. KEITHLEY <kkeithle@redhat.com>
 1666 Date:   Thu May 28 08:26:47 2020 -0400
 1667 
 1668     common-ha: ganesha-ha.sh bad test for {rhel,centos} for pcs options
 1669     
 1670     bash [[ ... =~ ... ]] built-in returns _0_ when the regex matches,
 1671     not 1, thus the sense of the test is backwards and never correctly
 1672     detects rhel or centos.
 1673     
 1674     master commit: eaf126f4b06a842977c1932ce699c4d76421a4b2
 1675     
 1676     Change-Id: Ic9e60aae4ea38aff8f13979080995e60621a68fe
 1677     Fixes: #1269
 1678     Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com>
 1679 
 1680 commit 1572f98486bf3204b5d2d89da880b92e26734c3c
 1681 Author: Pranith Kumar K <pkarampu@redhat.com>
 1682 Date:   Wed May 6 17:36:20 2020 +0530
 1683 
 1684     tests: skip tests on absence of reflink in xfs
 1685     
 1686     Fixes: #1223
 1687     Change-Id: I36cb72d920ffd77405051546615c5262c392daef
 1688     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 1689     (cherry picked from commit b85f01abab658d1d704cd6caf84dd64eddafbff7)
 1690 
 1691 commit 2ae5b28030b08348ec5e4c08ada8600943db0366
 1692 Author: Csaba Henk <csaba@redhat.com>
 1693 Date:   Fri May 8 23:01:04 2020 +0200
 1694 
 1695     fuse: occasional logging for fuse device 'weird' write errors
 1696     
 1697     This change is a followup to
 1698     I510158843e4b1d482bdc496c2e97b1860dc1ba93.
 1699     
 1700     In referred change we pushed log messages about 'weird'
 1701     write errors to fuse device out of sight, by reporting
 1702     them at Debug loglevel instead of Error (where
 1703     'weird' means errno is not POSIX compliant but having
 1704     meaningful semantics for FUSE protocol).
 1705     
 1706     This solved the issue of spurious error reporting.
 1707     And so far so good: these messages don't indicate
 1708     an error condition by themselves. However, when they
 1709     come in high repetitions, that indicates a suboptimal
 1710     condition which should be reported.[1]
 1711     
 1712     Therefore now we shall emit a Warning if a certain
 1713     errno occurs a certain number of times[2] as the
 1714     outcome of a write to the fuse device.
 1715     
 1716     ___
 1717     [1] typically ENOENTs and ENOTDIRs accumulate
 1718     when glusterfs' inode invalidation lags behind
 1719     the kernel's internal inode garbage collection
 1720     (in this case above errnos mean that the inode
 1721     which we requested to be invalidated is not found
 1722     in kernel). This can be mitigated with the
 1723     invalidate-limit command line / mount option,
 1724     cf. bz#1732717.
 1725     
 1726     [2] 256, as of the current implementation.
 1727     
 1728     Change-Id: I8cc7fe104da43a88875f93b0db49d5677cc16045
 1729     Updates: #1000
 1730     Signed-off-by: Csaba Henk <csaba@redhat.com>
 1731     (cherry picked from commit c1baf3c68b87584aea5389af958326f6ed01d7ec)
 1732 
 1733 commit 984a8dd8fa3e371fa1336730d6dc0041ceef877c
 1734 Author: Ashish Pandey <aspandey@redhat.com>
 1735 Date:   Tue May 5 18:17:49 2020 +0530
 1736 
 1737     cluster/ec: Return correct error code and log message
 1738     
 1739     In case of readdir was send with an FD on which opendir
 1740     was failed, this FD will be useless and we return it with error.
 1741     For now, we are returning it with EINVAL without logging any
 1742     message in log file.
 1743     
 1744     Return a correct error code and also log the message to improve thing to debug.
 1745     
 1746     fixes: #1220
 1747     Change-Id: Iaf035254b9c5aa52fa43ace72d328be622b06169
 1748     (cherry picked from commit af70cb5eedd80207cd184e69f2a4fb252b72d070)
 1749 
 1750 commit 1c8af2522458e3763302ec408fe1832dff150efa
 1751 Author: Ravishankar N <ravishankar@redhat.com>
 1752 Date:   Fri Apr 10 14:30:57 2020 +0530
 1753 
 1754     afr: event gen changes
 1755     
 1756     The general idea of the changes is to prevent resetting event generation
 1757     to zero in the inode ctx, since event gen is something that should
 1758     follow 'causal order'.
 1759     
 1760     Change #1:
 1761     For a read txn, in inode refresh cbk, if event_generation is
 1762     found zero, we are failing the read fop. This is not needed
 1763     because change in event gen is only a marker for the next inode refresh to
 1764     happen and should not be taken into account by the current read txn.
 1765     
 1766     Change #2:
 1767     The event gen being zero above can happen if there is a racing lookup,
 1768     which resets even get (in afr_lookup_done) if there are non zero afr
 1769     xattrs. The resetting is done only to trigger an inode refresh and a
 1770     possible client side heal on the next lookup. That can be acheived by
 1771     setting the need_refresh flag in the inode ctx. So replaced all
 1772     occurences of resetting even gen to zero with a call to
 1773     afr_inode_need_refresh_set().
 1774     
 1775     Change #3:
 1776     In both lookup and discover path, we are doing an inode refresh which is
 1777     not required since all 3 essentially do the same thing- update the inode
 1778     ctx with the good/bad copies from the brick replies. Inode refresh also
 1779     triggers background heals, but I think it is okay to do it when we call
 1780     refresh during the read and write txns and not in the lookup path.
 1781     
 1782     The .ts which relied on inode refresh in lookup path to trigger heals are
 1783     now changed to do read txn so that inode refresh and the heal happens.
 1784     
 1785     Change-Id: Iebf39a9be6ffd7ffd6e4046c96b0fa78ade6c5ec
 1786     Fixes: #1179
 1787     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
 1788     Reported-by: Erik Jacobson <erik.jacobson at hpe.com>
 1789     (cherry picked from commit f0fcd909ad4535b60c9208d4804ebe6afe421a09)
 1790 
 1791 commit f949c3e13e42d52fd5c230fc2f2607d327aad9d1
 1792 Author: Rinku Kothiya <rkothiya@redhat.com>
 1793 Date:   Fri Apr 17 13:26:50 2020 +0000
 1794 
 1795     Update rfc.sh for release-8
 1796     
 1797     updates: #1180
 1798     
 1799     Signed-off-by: Rinku Kothiya <rkothiya@redhat.com>
 1800 
 1801 commit 0e94dbb81154edcc7d1182527663a7340fbeab01
 1802 Author: Barak Sason Rofman <sason922@gmail.com>
 1803 Date:   Sun Apr 12 18:19:41 2020 +0300
 1804 
 1805     dht - Remove "tier" code (part 1)
 1806     
 1807     This patch is removing some of the "tier" code in dht xlator, as it is no longer
 1808     being used.
 1809     Not all of the not-needed code is removed at once, so reviewing is easier.
 1810     Follow up patches removing additional unused code will follow.
 1811     
 1812     This is based in the work done in https://review.gluster.org/#/c/glusterfs/+/23935/
 1813     
 1814     Change-Id: I3cb6a0c5d8f14afcd87cf021ef8f74b91c0f908a
 1815     updates: #1097
 1816     Signed-off-by: Barak Sason Rofman <bsaonro@redhat.com>
 1817 
 1818 commit 39f4a03cbaa7b3b93a3750258607cbd91b9d4b9d
 1819 Author: Mohit Agrawal <moagrawal@redhat.com>
 1820 Date:   Fri Apr 10 13:45:50 2020 +0530
 1821 
 1822     tests: Fix for spurious failure for some test cases
 1823     
 1824     Problem: Sometimes test case is failing at the time of creating files
 1825              on mount point after mounting the volume
 1826     
 1827     Solution: After started the volume need to wait to make sure all
 1828               bricks instances are completely started so put a online_brick_count
 1829               check after just started the volume
 1830     
 1831     Change-Id: I5020e7e417539377277ca00189f9c51d2cf877a6
 1832     Fixes: #1162
 1833     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 1834 
 1835 commit f507c20bd11565c329d103075a15da11dba461d5
 1836 Author: Xavi Hernandez <xhernandez@redhat.com>
 1837 Date:   Mon Mar 30 11:09:39 2020 +0200
 1838 
 1839     md-cache: avoid clearing cache when not necessary
 1840     
 1841     mdc_inode_xatt_set() blindly cleared current cache when dict was not
 1842     NULL, even if there was no xattr requested.
 1843     
 1844     This patch fixes this by only calling mdc_inode_xatt_set() when we have
 1845     explicitly requested something to cache.
 1846     
 1847     Change-Id: Idc91a4693f1ff39f7059acde26682ccc361b947d
 1848     Fixes: #1140
 1849     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
 1850 
 1851 commit 453861088dfabde2c41bec674bed2a6886b4ed37
 1852 Author: Sanju Rakonde <srakonde@redhat.com>
 1853 Date:   Wed Dec 25 22:56:12 2019 +0530
 1854 
 1855     posix: fix GF_VALIDATE_OR_GOTO(this->name, this, out)
 1856     
 1857     Remove GF_VALIDATE_OR_GOTO(this->name, this, out) when this
 1858     is passed as an argument and is checked for NULL in the caller
 1859     itself.
 1860     
 1861     GF_VALIDATE_OR_GOTO(this->name, this, out) is modified to use
 1862     xlator name instead of this->name as we are still verifying
 1863     whether this is NULL.
 1864     
 1865     updates: #1000
 1866     
 1867     Change-Id: Ide3180da29d0d4a35b2c5b9a7604fdf2ff4a9ffb
 1868     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
 1869 
 1870 commit d0cb5cbaff22ec5fcb685a327d6d54e5ae8d87f2
 1871 Author: Mohit Agrawal <moagrawal@redhat.com>
 1872 Date:   Thu Apr 9 13:08:23 2020 +0530
 1873 
 1874     test: tests/bugs/rpc/bug-847624.t is crashed
 1875     
 1876     Problem: glusterfs(GNFS) is crashing at the time of handling
 1877              Pollerr event in rpcsvc_drc_client_unref.GNFS is crashed
 1878              because ref was 0 at the time of unref and ref was taken
 1879              while Pollin event successfully handled.
 1880     
 1881     Solution: Convert drc_client ref to atomic ref to avoid the crash
 1882     
 1883     Change-Id: Ia4c054f2f388032a5cd99597d0cfa18b003ca690
 1884     Fixes: #1038
 1885     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 1886 
 1887 commit 0407d10f147862eb1c7854aed3adfbc3e1503c5d
 1888 Author: Dmitry Antipov <dmantipov@yandex.ru>
 1889 Date:   Fri Apr 10 11:13:12 2020 +0300
 1890 
 1891     tests: do not truncate file offsets and sizes to 32-bit
 1892     
 1893     Do not truncate file offsets and sizes to 32-bit to
 1894     prevent tests from spurious failures on >2Gb files.
 1895     
 1896     Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru>
 1897     
 1898     Change-Id: I2a77ea5f9f415249b23035eecf07129f19194ac2
 1899     Fixes: #1161
 1900 
 1901 commit 0abdd69636c42ec410a0615763f5c2ca4dca8f75
 1902 Author: Kaleb S. KEITHLEY <kkeithle@redhat.com>
 1903 Date:   Tue Apr 14 07:59:22 2020 -0400
 1904 
 1905     common-ha: cluster status shows "FAILOVER" when actually HEALTHY
 1906     
 1907     pacemaker devs change the format of the ouput of `pcs status`
 1908     
 1909     Expected to find a line in the format:
 1910     
 1911     Online: ....
 1912     
 1913     but now it's
 1914     
 1915       * Online: ...
 1916     
 1917     And the `grep -E "^Online:" no longer finds the list of nodes that
 1918     are online.
 1919     
 1920     Change-Id: If2aa1e7b53c766c625d7b4cc222a83ea2c0bd72d
 1921     Fixes: #1169
 1922     Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com>
 1923 
 1924 commit 43ed6fa34b259b568ae1442b65a5004a894a2654
 1925 Author: Rinku Kothiya <rkothiya@redhat.com>
 1926 Date:   Thu Apr 9 13:51:22 2020 +0000
 1927 
 1928     tests: Fix spurious failure of worm.t
 1929     
 1930     When the output of date command is a single digit
 1931     number it is preceded by zero which is getting
 1932     considered as an octal number. Removing the leading
 1933     zero from the number solved the problem.
 1934     
 1935     Fixes: #1156
 1936     
 1937     Change-Id: Iac4fa20607c0bb90d94dd8ff157ef6b60932c560
 1938     Signed-off-by: Rinku Kothiya <rkothiya@redhat.com>
 1939 
 1940 commit 0644fce4e52cd738f7ebb5e0596481a256fed3da
 1941 Author: Dmitry Antipov <dmantipov@yandex.ru>
 1942 Date:   Fri Apr 10 10:47:17 2020 +0300
 1943 
 1944     mgmt/glusterd: use stat() syscall wrapper
 1945     
 1946     Found with 0-symbol-check.t:
 1947     
 1948     ./tests/basic/0symbol-check.t ..
 1949     1..2
 1950     ./xlators/mgmt/glusterd/src/.libs/glusterd_la-glusterd-volume-set.o should call sys_stat, not stat
 1951     ok   1 [     40/  41011] <  40> 'find . -name *.o -exec ./tests/basic/symbol-check.sh {} \;'
 1952     not ok   2 [     11/      1] <  42> '[ ! -e ./.symbol-check-errors ]' -> ''
 1953     Failed 1/2 subtests
 1954     
 1955     Change-Id: I8962f487cd88738a1f7a962049d513712687088c
 1956     Fixes: #1160
 1957     Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru>
 1958 
 1959 commit d357863f4fc5be09b96ef3b359100d741efc0b4f
 1960 Author: Barak Sason Rofman <sason922@gmail.com>
 1961 Date:   Sun Apr 12 13:50:33 2020 +0300
 1962 
 1963     test: Fix test "bug-1064148" to pass in mux
 1964     
 1965     Parts of the test weren't designed to run in mux mode, this is now fixed
 1966     
 1967     Change-Id: I428c2fcce6d047e324ca5dcaef677ee1794e3dfe
 1968     updates: #1154
 1969     Signed-off-by: Barak Sason Rofman <bsasonro@redhat.com>
 1970 
 1971 commit cd77bddacb4eefb8c447c82220780626b878c3f0
 1972 Author: Mohit Agrawal <moagrawal@redhat.com>
 1973 Date:   Thu Apr 9 21:55:46 2020 +0530
 1974 
 1975     tests: Fix spurious failure of tests/bugs/glusterd/serialize-shd-manager-glusterd-restart.t
 1976     
 1977     Problem: Sometime volume status is failed after restart glusterd
 1978              in one cluster node
 1979     
 1980     Solution: Wait to finish glusterd handshake on down cluster node
 1981     
 1982     Change-Id: Ib23ca41c943caf2903c61ebf42dc437c1b9d6054
 1983     Fixes: #1158
 1984     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 1985 
 1986 commit a57980e37416994b99b20d19a495527fb49c0372
 1987 Author: Sheetal Pamecha <spamecha@redhat.com>
 1988 Date:   Thu Apr 9 01:59:03 2020 +0530
 1989 
 1990     protocol/server: Fix coverity issue RESOURCE_LEAK
 1991     
 1992     Handle case of arg not freed
 1993     CID: 1422174
 1994     
 1995     Updates: #1060
 1996     Change-Id: Ibd03908a3ea8369035c2b7f6e024b3e5be48f436
 1997     Signed-off-by: Sheetal Pamecha <spamecha@redhat.com>
 1998 
 1999 commit 82db4f23d201b1078129d083370563f68251d897
 2000 Author: Shwetha K Acharya <sacharya@redhat.com>
 2001 Date:   Tue Jan 14 08:26:12 2020 +0530
 2002 
 2003     Adding basic glusterfind test case
 2004     
 2005     This test case includes all the basic glusterfind scenarios.
 2006     
 2007     fixes: #1044
 2008     Change-Id: I6021443729e35769fe855c5cc41bb3fbc6365ef0
 2009     Signed-off-by: Shwetha K Acharya <sacharya@redhat.com>
 2010 
 2011 commit 18f70e806bcef5a3680462a4dbb7061e5eceee1c
 2012 Author: Mohit Agrawal <moagrawal@redhat.com>
 2013 Date:   Mon Apr 6 21:58:03 2020 +0530
 2014 
 2015     posix: Avoid dict_del logs in posix_is_layout_stale while key is NULL
 2016     
 2017     Problem: The key "GF_PREOP_PARENT_KEY" has been populated by dht and
 2018              for non-distribute volume like 1x3 key is not populated so
 2019              posix_is_layout stale throw a message while a file is created
 2020     
 2021     Solution: To avoid a log put a condition before delete a key
 2022     
 2023     Change-Id: I813ee7960633e7f9f5e9ad2f42f288053d9eb71f
 2024     Fixes: #1150
 2025     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 2026 
 2027 commit 9115dd39e953953495d48c6d6f24f44828f8bd5f
 2028 Author: Pranith Kumar K <pkarampu@redhat.com>
 2029 Date:   Tue Apr 7 16:05:16 2020 +0530
 2030 
 2031     mount/fuse: Wait for 'mount' child to exit before dying
 2032     
 2033     Problem:
 2034     tests/bugs/protocol/bug-1433815-auth-allow.t fails
 2035     sometimes because of stale mount. This stale mount
 2036     comes into picture when parent process dies without
 2037     waiting for the child process which mounts fuse fs
 2038     to die
 2039     
 2040     Fix:
 2041     Wait for mounting child process to die before dying.
 2042     
 2043     Fixes: #1152
 2044     Change-Id: I8baee8720e88614fdb762ea822d5877973eef8dc
 2045     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 2046 
 2047 commit 4398f68dada29418d2ec68ec538ec1378108221a
 2048 Author: kshithijiyer <kshithij.ki@gmail.com>
 2049 Date:   Fri Mar 27 11:13:29 2020 +0530
 2050 
 2051     [geo-rep] Merging multiple import into single import
 2052     
 2053     Geo-replication has a large number of repeated imports as shown below:
 2054     ```
 2055     from syncdutils import set_term_handler, finalize, lf
 2056     from syncdutils import log_raise_exception, FreeObject, escape
 2057     ```
 2058     
 2059     There imports can be clubbed together as shown below:
 2060     ``
 2061     from syncdutils import (set_term_handler, finalize, lf,
 2062                             log_raise_exception, FreeObject, escape)
 2063     ```
 2064     
 2065     Fixes: #1105
 2066     Change-Id: I59a48dd57a70fc851d93150b85e736ce41e8b793
 2067     Signed-off-by: kshithijiyer <kshithij.ki@gmail.com>
 2068 
 2069 commit 6917f52e6a9fd65f30f81302f3b84c734950810d
 2070 Author: Barak Sason Rofman <bsasonro@redhat.com>
 2071 Date:   Wed Jan 15 12:02:05 2020 +0200
 2072 
 2073     dht - fixing a permission update issue
 2074     
 2075     When bringing back a downed brick and performing lookup from the client
 2076     side, the permission on said brick aren't updated on the first lookup,
 2077     but only on the second.
 2078     
 2079     This patch modifies permission update logic so the first lookup will
 2080     trigger a permission update on the downed brick.
 2081     
 2082     LIMITATIONS OF THE PATCH:
 2083     As the choice of source depends on whether the directory has layout or not.
 2084     Even the directories on the newly added brick will have layout xattr[zeroed], but the same is not true for a root directory.
 2085     Hence, in case in the entire cluster only the newly added bricks are up [and others are down], then any change in permission during this time will be overwritten by the older permissions when the cluster is restarted.
 2086     
 2087     fixes: #999
 2088     Change-Id: Ieb70246d41e59f9cae9f70bc203627a433dfbd33
 2089     Signed-off-by: Barak Sason Rofman <bsasonro@redhat.com>
 2090 
 2091 commit ae8867fcfa5b14008a1c9a29674e971f502e1a61
 2092 Author: Pranith Kumar K <pkarampu@redhat.com>
 2093 Date:   Tue Apr 7 20:55:19 2020 +0530
 2094 
 2095     tests: Fix spurious failure of tests/bugs/snapshot/bug-1111041.t
 2096     
 2097     Test should wait for process down notification to be received
 2098     by glusterd.
 2099     
 2100     Fixes: #1153
 2101     Change-Id: I9162b58a92c1a909ca98097f14c0714f9086bdd1
 2102     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 2103 
 2104 commit 55914f968d907ed747774da15285b42653afda61
 2105 Author: Soumya Koduri <skoduri@redhat.com>
 2106 Date:   Mon Apr 6 12:36:44 2020 +0530
 2107 
 2108     gfapi: Suspend synctasks instead of blocking them
 2109     
 2110     There are certain conditions which blocks the current
 2111     execution thread (like waiting on mutex lock or condition
 2112     variable or I/O response). In such cases, if it is a
 2113     synctask thread, we should suspend the task instead
 2114     of blocking it (like done in SYNCOP using synctask_yield)
 2115     
 2116     This is to avoid deadlock like the one mentioned below -
 2117     
 2118     1) synctaskA sets fs->migration_in_progress to 1 and
 2119        does I/O (LOOKUP)
 2120     2) Other synctask threads wait for fs->migration_in_progress
 2121       to be reset to 0 by synctaskA and hence blocked
 2122     3) but synctaskA cannot resume as all synctask threads are blocked
 2123        on (2).
 2124     
 2125     Note: this same approach is already used by few other components
 2126     like syncbarrier etc.
 2127     
 2128     Change-Id: If90f870d663bb242c702a5b86ac52eeda67c6f0d
 2129     Fixes: #1146
 2130     Signed-off-by: Soumya Koduri <skoduri@redhat.com>
 2131 
 2132 commit e1ad1483e0e82d70fcc32311b7842bc9a77c04aa
 2133 Author: karthik-us <ksubrahm@redhat.com>
 2134 Date:   Thu Apr 2 13:03:35 2020 +0530
 2135 
 2136     cluster/afr: Removing unsupported options from code base to improve coverage
 2137     
 2138     Support for gluster volume heal <volname> info healed/heal-failed
 2139     was removed by commit bb02cfb56ae08f56df4452c2b948fa962ae1212b in
 2140     release-3.6. cli parser will display the usage message in all the
 2141     supported versions whenever these clis are run, leading to some
 2142     dead code in the latest branches. Since support for these clis
 2143     were removed long back, this should not give any backward
 2144     compatibility issues as well. Hence removing the dead code from
 2145     the code base which will lead to better code coverage by the
 2146     regression runs as well.
 2147     
 2148     Updates: #1052
 2149     Change-Id: I0c2b061469caf233c06d9699b0d159ce48e240b9
 2150     Signed-off-by: karthik-us <ksubrahm@redhat.com>
 2151 
 2152 commit f2d90013aaa4652dbcc1e6e3d62a0d31ada89d5d
 2153 Author: Mohit Agrawal <moagrawal@redhat.com>
 2154 Date:   Thu Dec 19 08:32:19 2019 +0530
 2155 
 2156     Posix: Optimize posix code to improve file creation
 2157     
 2158     Problem: Before executing a fop in POSIX xlator it builds an internal
 2159              path based on GFID.To validate the path it call's (l)stat
 2160              system call and while .glusterfs is heavily loaded kernel takes
 2161              time to lookup inode and due to that performance drops
 2162     
 2163     Solution: In this patch we followed two ways to improve the performance.
 2164               1) Keep open fd specific to first level directory(gfid[0])
 2165                  in .glusterfs, it would force to kernel keep the inodes
 2166                  from all those files in cache. In case of memory pressure
 2167                  kernel won't uncache first level inodes. We need to open
 2168                  256 fd's per brick to access the entry faster.
 2169               2) Use at based call's to access relative path to reduce
 2170                  path based lookup time.
 2171     
 2172     Note: To verify the patch we have executed kernel untar 100 times on 6
 2173           different clients after enabling metadata group-cache and some
 2174           other option.We were getting more than 20 percent improvement in
 2175           kenel untar after applying the patch.
 2176     
 2177     Credits: Xavi Hernandez <xhernandez@redhat.com>
 2178     Change-Id: I1643e6b01ed669b2bb148d02f4e6a8e08da45343
 2179     updates: #891
 2180     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 2181 
 2182 commit 28299f7852229be333102e08aae9b7fd4ad3338a
 2183 Author: Ashish Pandey <aspandey@redhat.com>
 2184 Date:   Wed Apr 1 13:31:50 2020 +0530
 2185 
 2186     cluster/ec: Add test for reset-brick command
 2187     
 2188     Following tests are done -
 2189     
 2190     1 - After finishing reset-brick all the bricks should be up.
 2191     2 - Heal should be completed.
 2192     3 - Check number of entries present on brick which was reset.
 2193     
 2194     Change-Id: I9314bed180293a99d400d94bb8cc7ece999da29e
 2195     Updates: #1144
 2196 
 2197 commit c61171ea24873b492d0100e376365db59235d321
 2198 Author: Vijay Bellur <vbellur@redhat.com>
 2199 Date:   Fri Apr 3 12:34:07 2020 -0700
 2200 
 2201     mgmt/glusterd: Reduce log level of repetitive log
 2202     
 2203     Noticed that the following message repeats quite a bit in log files when
 2204     an external monitoring tool queries gluster for list of volumes
 2205     periodically:
 2206     
 2207     "Received get vol req"
 2208     
 2209     As there's not much value in having this log message at log level INFO,
 2210     changing the log level to DEBUG to make glusterd.log a bit quieter.
 2211     
 2212     Change-Id: I4e791fc65b9a4f813d295e7b2b6a05f3c0782e69
 2213     Updates: #1000
 2214     Signed-off-by: Vijay Bellur <vbellur@redhat.com>
 2215 
 2216 commit 945b0a12b7a47039260330e02b89919c2cefb8d2
 2217 Author: Pranith Kumar K <pkarampu@redhat.com>
 2218 Date:   Fri Apr 3 15:25:44 2020 +0530
 2219 
 2220     fuse: Add error-logs to debug bug-1433815-auth-allow.t failures
 2221     
 2222     Fixes: #1149
 2223     Change-Id: I38483fc7d76d7fe0ac9fb649669a46bdf9c82234
 2224     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 2225 
 2226 commit 208b6d7b1a4d98e62d4bfa3b342bb78ade078799
 2227 Author: Raghavendra Bhat <raghavendra@redhat.com>
 2228 Date:   Tue Mar 31 11:25:51 2020 -0400
 2229 
 2230     storage/posix: log the ENOENT errors in posix_pstat
 2231     
 2232     Change-Id: I93f11dae6e4939ab79b0481ead2a4f7bb3085b70
 2233     Fixes: #1142
 2234     Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
 2235 
 2236 commit 77889c34b6f169dfc3257e2b566afa2f4bad1697
 2237 Author: Xavi Hernandez <xhernandez@redhat.com>
 2238 Date:   Fri Mar 27 23:56:15 2020 +0100
 2239 
 2240     write-behind: fix data corruption
 2241     
 2242     There was a bug in write-behind that allowed a previous completed write
 2243     to overwrite the overlapping region of data from a future write.
 2244     
 2245     Suppose we want to send three writes (W1, W2 and W3). W1 and W2 are
 2246     sequential, and W3 writes at the same offset of W2:
 2247     
 2248         W2.offset = W3.offset = W1.offset + W1.size
 2249     
 2250     Both W1 and W2 are sent in parallel. W3 is only sent after W2 completes.
 2251     So W3 should *always* overwrite the overlapping part of W2.
 2252     
 2253     Suppose write-behind processes the requests from 2 concurrent threads:
 2254     
 2255         Thread 1                    Thread 2
 2256     
 2257         <received W1>
 2258                                     <received W2>
 2259         wb_enqueue_tempted(W1)
 2260         /* W1 is assigned gen X */
 2261                                     wb_enqueue_tempted(W2)
 2262                                     /* W2 is assigned gen X */
 2263     
 2264                                     wb_process_queue()
 2265                                       __wb_preprocess_winds()
 2266                                         /* W1 and W2 are sequential and all
 2267                                          * other requisites are met to merge
 2268                                          * both requests. */
 2269                                         __wb_collapse_small_writes(W1, W2)
 2270                                         __wb_fulfill_request(W2)
 2271     
 2272                                       __wb_pick_unwinds() -> W2
 2273                                       /* In this case, since the request is
 2274                                        * already fulfilled, wb_inode->gen
 2275                                        * is not updated. */
 2276     
 2277                                     wb_do_unwinds()
 2278                                       STACK_UNWIND(W2)
 2279     
 2280                                     /* The application has received the
 2281                                      * result of W2, so it can send W3. */
 2282                                     <received W3>
 2283     
 2284                                     wb_enqueue_tempted(W3)
 2285                                     /* W3 is assigned gen X */
 2286     
 2287                                     wb_process_queue()
 2288                                       /* Here we have W1 (which contains
 2289                                        * the conflicting W2) and W3 with
 2290                                        * same gen, so they are interpreted
 2291                                        * as concurrent writes that do not
 2292                                        * conflict. */
 2293                                       __wb_pick_winds() -> W3
 2294     
 2295                                     wb_do_winds()
 2296                                       STACK_WIND(W3)
 2297     
 2298         wb_process_queue()
 2299           /* Eventually W1 will be
 2300            * ready to be sent */
 2301           __wb_pick_winds() -> W1
 2302           __wb_pick_unwinds() -> W1
 2303             /* Here wb_inode->gen is
 2304              * incremented. */
 2305     
 2306         wb_do_unwinds()
 2307           STACK_UNWIND(W1)
 2308     
 2309         wb_do_winds()
 2310           STACK_WIND(W1)
 2311     
 2312     So, as we can see, W3 is sent before W1, which shouldn't happen.
 2313     
 2314     The problem is that wb_inode->gen is only incremented for requests that
 2315     have not been fulfilled but, after a merge, the request is marked as
 2316     fulfilled even though it has not been sent to the brick. This allows
 2317     that future requests are assigned to the same generation, which could
 2318     be internally reordered.
 2319     
 2320     Solution:
 2321     
 2322     Increment wb_inode->gen before any unwind, even if it's for a fulfilled
 2323     request.
 2324     
 2325     Special thanks to Stefan Ring for writing a reproducer that has been
 2326     crucial to identify the issue.
 2327     
 2328     Change-Id: Id4ab0f294a09aca9a863ecaeef8856474662ab45
 2329     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
 2330     Fixes: #884
 2331 
 2332 commit f55c45537435cc2a1412c80a6c922db63ce65965
 2333 Author: Pranith Kumar K <pkarampu@redhat.com>
 2334 Date:   Thu Apr 2 15:30:28 2020 +0530
 2335 
 2336     features/utime: Don't access frame after stack-wind
 2337     
 2338     Problem:
 2339     frame is accessed after stack-wind. This can lead to crash
 2340     if the cbk frees the frame.
 2341     
 2342     Fix:
 2343     Use new frame for the wind instead.
 2344     
 2345     Updates: #832
 2346     Change-Id: I64754609f1114b0bbd4d1336fa81a56f2cca6e03
 2347     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 2348 
 2349 commit a7d7ec066e56ac03bf252c26beb20fdc2c3b6772
 2350 Author: Sunny Kumar <sunkumar@redhat.com>
 2351 Date:   Thu Mar 26 10:46:16 2020 +0000
 2352 
 2353     snap_scheduler: python3 compatibility and new test case
 2354     
 2355     Problem:
 2356     "snap_scheduler.py init" command failing with the below traceback:
 2357     
 2358     [root@dhcp43-104 ~]# snap_scheduler.py init
 2359     Traceback (most recent call last):
 2360       File "/usr/sbin/snap_scheduler.py", line 941, in <module>
 2361         sys.exit(main(sys.argv[1:]))
 2362       File "/usr/sbin/snap_scheduler.py", line 851, in main
 2363         initLogger()
 2364       File "/usr/sbin/snap_scheduler.py", line 153, in initLogger
 2365         logfile = os.path.join(process.stdout.read()[:-1], SCRIPT_NAME + ".log")
 2366       File "/usr/lib64/python3.6/posixpath.py", line 94, in join
 2367         genericpath._check_arg_types('join', a, *p)
 2368       File "/usr/lib64/python3.6/genericpath.py", line 151, in _check_arg_types
 2369         raise TypeError("Can't mix strings and bytes in path components") from None
 2370     TypeError: Can't mix strings and bytes in path components
 2371     
 2372     Solution:
 2373     
 2374     Added the 'universal_newlines' flag to Popen to support backward compatibility.
 2375     
 2376     Added a basic test for snapshot scheduler.
 2377     
 2378     Change-Id: I78e8fabd866fd96638747ecd21d292f5ca074a4e
 2379     Fixes: #1134
 2380     Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
 2381 
 2382 commit 80dd8cceab3b860bf1bc2945c8e2d8d0b3913e48
 2383 Author: Mohit Agrawal <moagrawal@redhat.com>
 2384 Date:   Tue Mar 31 16:45:35 2020 +0530
 2385 
 2386     rpc: Make ssl log more useful
 2387     
 2388     Currently, ssl_setup_connection_params throws 4 messages for every
 2389     rpc connection that irritates a user while reading the logs. The same
 2390     info we can print in a single log with peerinfo to make it more
 2391     useful.ssl_setup_connection_params try to load dh_param even user
 2392     has not configured it and if a dh_param file is not available it throws
 2393     a failure message.To avoid the message load dh_param only while the user
 2394     has configured it.
 2395     
 2396     Change-Id: I9ddb57f86a3fa3e519180cb5d88828e59fe0e487
 2397     Fixes: #1141
 2398     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 2399 
 2400 commit 2d5ba449e9200b16184b1e7fc84cabd015f1f779
 2401 Author: Ravishankar N <ravishankar@redhat.com>
 2402 Date:   Mon Dec 24 13:00:19 2018 +0530
 2403 
 2404     afr: mark pending xattrs as a part of metadata heal
 2405     
 2406     ...if pending xattrs are zero for all children.
 2407     
 2408     Problem:
 2409     If there are no pending xattrs and a metadata heal needs to be
 2410     performed, it can be possible that we end up with xattrs inadvertendly
 2411     deleted from all bricks, as explained in the  BZ.
 2412     
 2413     Fix:
 2414     After picking one among the sources as the good copy, mark pending xattrs on
 2415     all sources to blame the sinks. Now even if this metadata heal fails midway,
 2416     a subsequent heal will still choose one of the valid sources that it
 2417     picked previously.
 2418     
 2419     Fixes: #1067
 2420     Change-Id: If1b050b70b0ad911e162c04db4d89b263e2b8d7b
 2421     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
 2422 
 2423 commit 6ef82683064a87c434bd67e2004935281e4c04cc
 2424 Author: l17zhou <cynthia.zhou@nokia-sbell.com>
 2425 Date:   Tue Dec 3 07:43:35 2019 +0200
 2426 
 2427     add clean local after grant lock
 2428     
 2429     found by flock test, without correct ref number of fd,
 2430     lock will not be correctly released.
 2431     
 2432     Fixes: bz#1779089
 2433     Change-Id: I3e466b17c852eb219c8778e43af8ad670a8449cc
 2434     Signed-off-by: l17zhou <cynthia.zhou@nokia-sbell.com>
 2435 
 2436 commit 936f8b2fc666a2c18b16bee7ab14da6dd7bfd5f2
 2437 Author: Hari Gowtham <hgowtham@redhat.com>
 2438 Date:   Mon Mar 30 14:08:09 2020 +0530
 2439 
 2440     Marker: Logically deadcode found by coverity
 2441     
 2442     Have removed the deadcode found by the coverity id:1356503
 2443     
 2444     Change-Id: Ieaa41e864538fb82dc967b4a214d4db09e267098
 2445     Updates: #1060
 2446     Signed-off-by: Hari Gowtham <hgowtham@redhat.com>
 2447 
 2448 commit e589d8de66d3325da8fbbbe44d1a5bd6335e08ab
 2449 Author: Mohit Agrawal <moagrawal@redhat.com>
 2450 Date:   Wed Mar 4 09:17:26 2020 +0530
 2451 
 2452     core[brick_mux]: brick crashed when creating and deleting volumes over time
 2453     
 2454     Problem: In brick_mux environment, while volumes are created/stopped in a loop
 2455              after running a long time the main brick is crashed.The brick is crashed
 2456              because the main brick process was not cleaned up memory for all objects
 2457              at the time of detaching a volume.
 2458              Below are the objects that are missed at the time of detaching a volume
 2459              1) xlator object for a brick graph
 2460              2) local_pool for posix_lock xlator
 2461              3) rpc object cleanup at quota xlator
 2462              4) inode leak at brick xlator
 2463     
 2464     Solution: To avoid the crash resolve all leak at the time of detaching a brick
 2465     Change-Id: Ibb6e46c5fba22b9441a88cbaf6b3278823235913
 2466     updates: #977
 2467     Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
 2468 
 2469 commit cc43ac8651de9aa508b01cb259b43c02d89b2afc
 2470 Author: Krutika Dhananjay <kdhananj@redhat.com>
 2471 Date:   Mon Mar 23 11:47:10 2020 +0530
 2472 
 2473     features/shard: Fix crash during shards cleanup in error cases
 2474     
 2475     A crash is seen during a reattempt to clean up shards in background
 2476     upon remount. And this happens even on remount (which means a remount
 2477     is no workaround for the crash).
 2478     
 2479     In such a situation, the in-memory base inode object will not be
 2480     existent (new process, non-existent base shard).
 2481     So local->resolver_base_inode will be NULL.
 2482     
 2483     In the event of an error (in this case, of space running out), the
 2484     process would crash at the time of logging the error in the following line -
 2485     
 2486             gf_msg(this->name, GF_LOG_ERROR, local->op_errno, SHARD_MSG_FOP_FAILED,
 2487                    "failed to delete shards of %s",
 2488                    uuid_utoa(local->resolver_base_inode->gfid));
 2489     
 2490     Fixed that by using local->base_gfid as the source of gfid when
 2491     local->resolver_base_inode is NULL.
 2492     
 2493     Change-Id: I0b49f2b58becd0d8874b3d4b14ff8d92a89d02d5
 2494     Fixes: #1127
 2495     Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
 2496 
 2497 commit c375f25b3e305cef2ab020866a10f26a7fbf1957
 2498 Author: Susant Palai <spalai@redhat.com>
 2499 Date:   Mon Mar 16 11:39:56 2020 +0530
 2500 
 2501     dht: gf_defrag_process_dir is called even if gf_defrag_fix_layout has failed
 2502     
 2503     Currently even though gf_defrag_fix_layout fails with ENOENT or ESTALE, a
 2504     subsequent call is made to gf_defrag_process_dir leading to rebalance failure.
 2505     
 2506     fixes: #1102
 2507     Change-Id: Ib0c309fd78e89a000fed3feb4bbe2c5b48e61478
 2508     Signed-off-by: Susant Palai <spalai@redhat.com>
 2509 
 2510 commit 7ee3c9a42fc7584d90da6158ef7510551a915b2b
 2511 Author: Sunny Kumar <sunkumar@redhat.com>
 2512 Date:   Tue Mar 17 15:42:59 2020 +0000
 2513 
 2514     worm-xlator: fix newly introduced coverity issue
 2515     
 2516     This patch fixes CID: 1420405
 2517     
 2518     updates: #1060
 2519     
 2520     Change-Id: I0524e999fa1d36ed5a713eabf65482c04ad43a1a
 2521     Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
 2522 
 2523 commit fb20713b380e1df8d7f9e9df96563be2f9144fd6
 2524 Author: Mohit Agrawal <moagrawal@redhat.com>
 2525 Date:   Thu Mar 12 21:12:13 2020 +0530
 2526 
 2527     Posix: Use simple approach to close fd
 2528     
 2529     Problem: posix_release(dir) functions add the fd's into a ctx->janitor_fds
 2530              and janitor thread closes the fd's.In brick_mux environment it is
 2531              difficult to handle race condition in janitor threads because brick
 2532              spawns a single janitor thread for all bricks.
 2533     
 2534     Solution: Use synctask to execute posix_release(dir) functions instead of
 2535               using background a thread to close fds.
 2536     
 2537     Credits: Pranith Karampuri <pkarampu@redhat.com>
 2538     Change-Id: Iffb031f0695a7da83d5a2f6bac8863dad225317e
 2539     Fixes: bz#1811631
 2540     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 2541 
 2542 commit 5410cc35ea09b1932c8eae4933fbf84f462e61ea
 2543 Author: Sunny Kumar <sunkumar@redhat.com>
 2544 Date:   Mon Mar 16 15:17:23 2020 +0000
 2545 
 2546     geo-rep: descriptive message when worker crashes due to EIO
 2547     
 2548     With this patch now you can notice log if it is due to EIO:
 2549     
 2550     [2020-03-16 16:24:48.293837] E [syncdutils(worker /bricks/brick1/mbr3):348:log_raise_exception] <top>: Getting "Input/Output error" is most likely due to a. Brick is down or b. Split brain issue.
 2551     [2020-03-16 16:24:48.293915] E [syncdutils(worker /bricks/brick1/mbr3):352:log_raise_exception] <top>: This is expected as per design to keep the consistency of the file system. Once the above issue is resolved geo-rep would automatically proceed further.
 2552     
 2553     Change-Id: Ie33f2440bc96089731ce12afa8dab91d9550a7ca
 2554     Fixes: #1104
 2555     Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
 2556 
 2557 commit 476139c962c298b2d00a3d42b90632fc25401526
 2558 Author: David Spisla <david.spisla@iternity.com>
 2559 Date:   Tue Feb 25 10:57:36 2020 +0100
 2560 
 2561     WORM-Xlator: Initial write of a file succeeds if auto-commit-period 0
 2562     
 2563     If worm-file-level enabled and auto-commit-period 0 an initial write
 2564     of a file (e.g. $ echo test >> file1.txt) would lead to an zero byte
 2565     file because the WORM xlator immediately WORMed the file when it was
 2566     created.
 2567     
 2568     To avoid this we move the setting of trusted.worm_file from
 2569     worm_create_cbk to worm_release . This means that this xattr will set
 2570     when the filehandle is closed and all initial WRITE FOPs succeed.
 2571     
 2572     Finally we also perform gf_worm_state_transition in worm_release to
 2573     ensure that the file will be immediately WORMed after the file handle
 2574     was closed.
 2575     
 2576     Change-Id: I5d02e18975b646ca1a27ed41d836e9d0dc333204
 2577     Fixes: bz#1808421
 2578     Signed-off-by: David Spisla <david.spisla@iternity.com>
 2579 
 2580 commit b30f76980fc5d01b569d9409b68d24d809444ff8
 2581 Author: Xavi Hernandez <xhernandez@redhat.com>
 2582 Date:   Sun Mar 8 18:36:45 2020 +0100
 2583 
 2584     open-behind: fix missing fd reference
 2585     
 2586     Open behind was not keeping any reference on fd's pending to be
 2587     opened. This makes it possible that a concurrent close and en entry
 2588     fop (unlink, rename, ...) caused destruction of the fd while it
 2589     was still being used.
 2590     
 2591     Change-Id: Ie9e992902cf2cd7be4af1f8b4e57af9bd6afd8e9
 2592     Fixes: bz#1810934
 2593     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
 2594 
 2595 commit 2e3e7a2309368376855ffac28620ad56f2c6b7b5
 2596 Author: Ravishankar N <ravishankar@redhat.com>
 2597 Date:   Fri Feb 21 18:13:57 2020 +0530
 2598 
 2599     tests: fix afr-lock-heal-* failure
 2600     
 2601     When brick-mux is enabled:
 2602     
 2603     i)brick statedumps seem to be listing the same lock information multiple times.
 2604     While that is getting fixed, make changes to the .ts to check for unique values.
 2605     
 2606     ii)detecting a brick as online via brick_up_status() seems to be taking
 2607     longer time when delaygen is enabled. Hence bump up PROCESS_UP_TIMEOUT to
 2608     90 for afr-lock-heal-advanced.t
 2609     
 2610     Updates: #1042
 2611     Change-Id: Ife76008f7a99dd1f1fe5791a32577366baaab4b3
 2612     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
 2613 
 2614 commit b164a74884becef281b57ef93428bb740e3e342e
 2615 Author: Pranith Kumar K <pkarampu@redhat.com>
 2616 Date:   Tue Feb 4 18:42:33 2020 +0530
 2617 
 2618     cluster/afr: Fixes for halo
 2619     
 2620     Current implementation assumes that ping-event will come after connect event
 2621     but that may not be the case in the cases where after socket connection fds
 2622     need to be re-opened which would consume more time. So handle any order of the
 2623     ping/child-up events.
 2624     
 2625     fixes: bz#1800583
 2626     Change-Id: I6bcdc0caa503bdc039ef2b4739fbf4afae121f05
 2627     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 2628 
 2629 commit eb916c057036db8289b41265797e5dce066d1512
 2630 Author: Amar Tumballi <amar@kadalu.io>
 2631 Date:   Wed Mar 4 00:46:08 2020 +0530
 2632 
 2633     utime: resolve an issue of permission denied logs
 2634     
 2635     In case where uid is not set to be 0, there are possible errors
 2636     from acl xlator. So, set `uid = 0;` with pid indicating this is
 2637     set from UTIME activity.
 2638     
 2639     The message "E [MSGID: 148002] [utime.c:146:gf_utime_set_mdata_setxattr_cbk] 0-dev_SNIP_data-utime: dict set of key for set-ctime-mdata failed [Permission denied]" repeated 2 times between [2019-12-19 21:27:55.042634] and [2019-12-19 21:27:55.047887]
 2640     
 2641     Change-Id: Ieadf329835a40a13ac0bf908dac776e66954466c
 2642     Fixes: #832
 2643     Signed-off-by: Amar Tumballi <amar@kadalu.io>
 2644 
 2645 commit 2668e9f1892dbe898e41e3a14701258d78cb8f46
 2646 Author: Sanju Rakonde <srakonde@redhat.com>
 2647 Date:   Tue Mar 10 15:05:08 2020 +0530
 2648 
 2649     cli: display the error while probing the localhost
 2650     
 2651     With bc6e206c6, we regressed in displaying the error
 2652     message when a user tries to probe localhost. With
 2653     this change, we display "probe on localhost not needed"
 2654     message to the user.
 2655     
 2656     credits: Sachin Prabhu <sprabhu@redhat.com>
 2657     
 2658     fixes: bz#1810042
 2659     
 2660     Change-Id: Ibf82b5a658c371c08290a0b4f655e5ac5f436c06
 2661     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
 2662 
 2663 commit 75691851391694d27ad5dcb797df8d4a82fd9e2d
 2664 Author: Mohammed Rafi KC <rkavunga@redhat.com>
 2665 Date:   Tue Mar 10 21:06:20 2020 +0530
 2666 
 2667     glusterd/snapshot: Add a warning message when a volume config changes
 2668     
 2669     While changing a volume configuration, there is a chance that the
 2670     brick layout on the disk might be changed. If snapshot is present
 2671     on such volumes that will effects it's working. So this patch adds
 2672     a warning message if snapshot is present while a volume config change
 2673     happen.
 2674     
 2675     Change-Id: I7256863fef734841fce0bc9ad94d5d201b1813d5
 2676     Fixes: bz#1812144
 2677     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
 2678 
 2679 commit 131d24d0c06ca82015256b98695421e09d27304a
 2680 Author: Barak Sason Rofman <bsasonro@redhat.com>
 2681 Date:   Sun Mar 8 17:27:47 2020 +0200
 2682 
 2683     dht - selfheal code cleaning
 2684     
 2685     1 - Converted methods to static
 2686     2 - Removed unused code
 2687     
 2688     Change-Id: I49db3e28116da1c3c9ff0a33dcce7281bc3856f7
 2689     updates: bz#1193929
 2690     Signed-off-by: Barak Sason Rofman <bsasonro@redhat.com>
 2691 
 2692 commit f90866a092f0df6e312be25807bf2118e163407d
 2693 Author: Arjun Sharma <arjsharm@redhat.com>
 2694 Date:   Wed Mar 11 11:37:48 2020 +0530
 2695 
 2696     Updating create-ganesha-config script:
 2697     
 2698     Adding the Security_Label parameter for labelled nfs
 2699     
 2700     Change-Id: I26d332bc30c767093cfa5d6e63a3b0268fc8a60b
 2701     Fixes: bz#1812353
 2702     Signed-off-by: Arjun Sharma <arjsharm@redhat.com>
 2703 
 2704 commit 6c492274e923df1083b61bd46453e11171761632
 2705 Author: l17zhou <cynthia.zhou@nokia-sbell.com>
 2706 Date:   Fri Mar 6 03:54:02 2020 +0200
 2707 
 2708     protocol/client: fallback to anonymous fd for fsync
 2709     
 2710     Change-Id: I32f801206ce7fbd05aa693f44c2f140304f2e275
 2711     Fixes: bz#1810842
 2712 
 2713 commit c12a7d9bdc3887c4ed43f76a3c7886727a62d0bc
 2714 Author: Purna Pavan Chandra Aekkaladevi <paekkala@redhat.com>
 2715 Date:   Wed Oct 30 12:42:16 2019 +0530
 2716 
 2717     Posix: structs re-aligned and manual padding added
 2718     
 2719     All the structs present inside xlators/storage/posix have been re-aligned
 2720     into memory efficient way. Manual padding has been added to remove
 2721     compile time padding. This manual padding is for development and
 2722     debugging benefits. It lets to use -Wpadded option and raise warnings
 2723     from the remaining structs where compile time padding happens.
 2724     
 2725     Change-Id: Ie72c02810803eae29fca435c71aa131a1315b8a8
 2726     Updates: bz#1754448
 2727     Signed-off-by: Purna Pavan Chandra Aekkaladevi <paekkala@redhat.com>
 2728 
 2729 commit 547fa5bbe5b3438d981de50ac5b2497683a4d9e2
 2730 Author: yatipadia <ypadia@redhat.com>
 2731 Date:   Tue Dec 31 14:47:03 2019 +0530
 2732 
 2733     mgmt/glusterd: Adding validation for statedump path
 2734     
 2735     Description of problem:
 2736     server.statedump-path is the path where statedumps are stored,
 2737     by default it is /var/run/gluster. And can be set to any valid
 2738     directory path. It was observed that server.statedump-path was
 2739     also accepting file, non-existent file and non-existent paths
 2740     as well. And statedump command was successful even when
 2741     statedumps with all the invalid paths.
 2742     a. A file
 2743     b. A non-existent path
 2744     
 2745     Solution:
 2746     Added a validation function in gluster-volume-set.c which will
 2747     allow volume set to success if it's a valid directory
 2748     and in all other cases, volume set should fail.
 2749     
 2750     Fixes: bz#1787122
 2751     
 2752     Change-Id: Ia66e2b3d35f23efc5444c829928779a79d827b42
 2753     Signed-off-by: yatipadia <ypadia@redhat.com>
 2754     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
 2755 
 2756 commit 3a9254d4ed5452c6da4401abd9626db6345629b7
 2757 Author: Barak Sason Rofman <bsasonro@redhat.com>
 2758 Date:   Sun Feb 9 15:09:30 2020 +0200
 2759 
 2760     dht/rebalance - fixing failure occurace due to rebalance stop
 2761     
 2762     Probelm description:
 2763     When topping rebalance, the following error messages appear in the
 2764     rebalance log file:
 2765     [2020-01-28 14:31:42.452070] W [dht-rebalance.c:3447:gf_defrag_process_dir] 0-distrep-dht: Found error from gf_defrag_get_entry
 2766     [2020-01-28 14:31:42.452764] E [MSGID: 109111] [dht-rebalance.c:3971:gf_defrag_fix_layout] 0-distrep-dht: gf_defrag_process_dir failed for directory: /0/1/2/3/4/5/6/7/8/9/10/11/12/13/14/15/16/17/18/19/20/21/22/23/24/25/26/27/28/29/30/31
 2767     [2020-01-28 14:31:42.453498] E [MSGID: 109016] [dht-rebalance.c:3906:gf_defrag_fix_layout] 0-distrep-dht: Fix layout failed for /0/1/2/3/4/5/6/7/8/9/10/11/12/13/14/15/16/17/18/19/20/21/22/23/24/25/26/27/28/29/30
 2768     
 2769     In order to avoid seing these error messages, a modification to the
 2770     error handling mechanism has been made.
 2771     In addition, several log messages had been added in order to improve debugging efficiency
 2772     
 2773     fixes: bz#1800956
 2774     Change-Id: Ifc82dae79ab3da9fe22ee25088a2a6b855afcfcf
 2775     Signed-off-by: Barak Sason Rofman <bsasonro@redhat.com>
 2776 
 2777 commit 3e059ec1216535a7866bcb8c3d6dc3cc8047b940
 2778 Author: yatipadia <ypadia@redhat.com>
 2779 Date:   Wed Dec 11 15:58:46 2019 +0530
 2780 
 2781     xlator/dht-helper: structure logging
 2782     
 2783     convert gf_msg() to gf_smsg()
 2784     
 2785     Updates: #657
 2786     
 2787     Change-Id: Iab35ac89b7d7fb6fb0074fc61b11bf679c517c9d
 2788     Signed-off-by: yatipadia <ypadia@redhat.com>
 2789     Signed-off-by: yatip <ypadia@redhat.com>
 2790 
 2791 commit 48245df36c5dec40779355a36d7bb7689f9efe6d
 2792 Author: yatip <ypadia@redhat.com>
 2793 Date:   Mon Feb 3 17:36:00 2020 +0530
 2794 
 2795     protocol/client: structure logging
 2796     
 2797     convert gf_msg() to gf_smsg()
 2798     
 2799     Updates: #657
 2800     
 2801     Change-Id: I76a09cfd283bb4ec5c4358536da66547aaf0de31
 2802     Signed-off-by: yatip <ypadia@redhat.com>
 2803 
 2804 commit 1291667e26516d02e548207ba3df84c456a36b35
 2805 Author: yatip <ypadia@redhat.com>
 2806 Date:   Thu Feb 27 18:48:16 2020 +0530
 2807 
 2808     protocol/server-helpers: structure logging
 2809     
 2810     convert gf_msg() to gf_smsg()
 2811     
 2812     Updates: #657
 2813     
 2814     Change-Id: Ic7b38b646fa0932f7c1562467866137c4567e1f1
 2815     Signed-off-by: yatip <ypadia@redhat.com>
 2816 
 2817 commit f9249c6d79ef654ec2a8e87efc202ce1df642e31
 2818 Author: yatip <ypadia@redhat.com>
 2819 Date:   Thu Feb 27 14:34:56 2020 +0530
 2820 
 2821     protocol/server-rpc-fops_v2: structure logging
 2822     
 2823     convert gf_msg() to gf_smsg()
 2824     
 2825     Updates: #657
 2826     
 2827     Change-Id: I01146bcd06bca44faeca29da48fab1ee3fc51e00
 2828     Signed-off-by: yatip <ypadia@redhat.com>
 2829 
 2830 commit 1b0e1cfd7dcede3c4075094b6e14ef9af384a7ec
 2831 Author: yatip <ypadia@redhat.com>
 2832 Date:   Thu Feb 20 14:14:43 2020 +0530
 2833 
 2834     protocol/server: structure logging
 2835     
 2836     Convert all gf_msg() to gf_smsg()
 2837     
 2838     Updates: #657
 2839     Change-Id: Ic54b03f05e2766c87f50df0b3a66803b5519fad9
 2840     Signed-off-by: yatip <ypadia@redhat.com>
 2841 
 2842 commit d0ac63e462a71c267436d0bbc8fa638c39b9c69f
 2843 Author: Xavi Hernandez <xhernandez@redhat.com>
 2844 Date:   Sun Mar 1 19:49:04 2020 +0100
 2845 
 2846     cluster/afr: fix race when bricks come up
 2847     
 2848     The was a problem when self-heal was sending lookups at the same time
 2849     that one of the bricks was coming up. In this case there was a chance
 2850     that the number of 'up' bricks changes in the middle of sending the
 2851     requests to subvolumes which caused a discrepancy in the expected
 2852     number of replies and the actual number of sent requests.
 2853     
 2854     This discrepancy caused that AFR continued executing requests before
 2855     all requests were complete. Eventually, the frame of the pending
 2856     request was destroyed when the operation terminated, causing a use-
 2857     after-free issue when the answer was finally received.
 2858     
 2859     In theory the same thing could happen in the reverse way, i.e. AFR
 2860     tries to wait for more replies than sent requests, causing a hang.
 2861     
 2862     Change-Id: I7ed6108554ca379d532efb1a29b2de8085410b70
 2863     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
 2864     Fixes: bz#1808875
 2865 
 2866 commit 54250605210b91715b3585e2a677870241364ee5
 2867 Author: Amar Tumballi <amar@kadalu.io>
 2868 Date:   Tue Feb 4 22:46:38 2020 +0530
 2869 
 2870     github templates: Change to reflect the future workflow
 2871     
 2872     * Issue template changed to remove reference to bugzilla
 2873     * PR template removed.
 2874     
 2875     CONTRIBUTING made .md friendly (will be renamed once we shift review process)
 2876     
 2877     Updates: #824
 2878     Change-Id: Ifd6f93ad9b7be97d8eb950a141fe585d3e5ed021
 2879     Signed-off-by: Amar Tumballi <amar@kadalu.io>
 2880 
 2881 commit b8d6a9bfb5ac1580f8dca06916db626789606f05
 2882 Author: yatipadia <ypadia@redhat.com>
 2883 Date:   Wed Dec 11 14:29:50 2019 +0530
 2884 
 2885     xlator/dht-lock: structure logging
 2886     
 2887     convert gf_msg() to gf_smsg()
 2888     
 2889     Change-Id: If540ca921b1cd8ca75b92b3d72eb9eb61bdaaa10
 2890     Updates: #657
 2891     Signed-off-by: yatip <ypadia@redhat.com>
 2892 
 2893 commit 3ba0885b5be42efec01ebd42cc5fbd33dc1b2756
 2894 Author: kinsu <vpolakis@gmail.com>
 2895 Date:   Thu Sep 19 08:34:32 2019 +0000
 2896 
 2897     Segmentation fault occurs during truncate
 2898     
 2899     Problem:
 2900     Segmentation fault occurs when bricks are nearly full 100% and in
 2901     parallel truncate of a file is attempted (No space left on device).
 2902     Prerequicite is that performance xlators are activated
 2903     (read-ahead, write-behind etc)
 2904     while stack unwind of the frames following an error responce
 2905     from brick (No space left on device) frame->local includes a memory
 2906     location that is not allocated via mem_get but via calloc.
 2907     The destroyed frame is always ra_truncate_cbk winded from ra_ftruncate
 2908     and the inode ptr is copied to the frame local in the wb_ftruncate.
 2909     
 2910     Fix:
 2911     extra check is added for the pool ptr
 2912     
 2913     Change-Id: Ic5d3bd0ab7011e40b2811c6dece063b256e4d9d1
 2914     Fixes: bz#1797882
 2915     Signed-off-by: kinsu <vpolakis@gmail.com>
 2916 
 2917 commit a0e08901a87700f967c26f146f54e2301f5d0594
 2918 Author: Andreas Loibl <andreas.loibl@check24.de>
 2919 Date:   Thu Feb 20 14:05:35 2020 +0100
 2920 
 2921     mount.glusterfs: don't strip / from subdir-mounts
 2922     
 2923     Change-Id: Ib07c31dc6669aa9d9f84a21e6160237290ed9afb
 2924     Fixes: bz#1804786
 2925 
 2926 commit 45e81aae791da9d013aba2286af44826227c05ec
 2927 Author: Vishal Pandey <vpandey@redhat.com>
 2928 Date:   Tue Nov 19 11:39:22 2019 +0530
 2929 
 2930     glusterd: Brick process fails to come up with brickmux on
 2931     
 2932     Issue:
 2933     1- In a cluster of 3 Nodes N1, N2, N3. Create 3 volumes vol1,
 2934     vol2, vol3 with 3 bricks (one from each node)
 2935     2- Set cluster.brick-multiplex on
 2936     3- Start all 3 volumes
 2937     4- Check if all bricks on a node are running on same port
 2938     5- Kill N1
 2939     6- Set performance.readdir-ahead for volumes vol1, vol2, vol3
 2940     7- Bring N1 up and check volume status
 2941     8- All bricks processes not running on N1.
 2942     
 2943     Root Cause -
 2944     Since, There is a diff in volfile versions in N1 as compared
 2945     to N2 and N3 therefore glusterd_import_friend_volume() is called.
 2946     glusterd_import_friend_volume() copies the new_volinfo and deletes
 2947     old_volinfo and then calls glusterd_start_bricks().
 2948     glusterd_start_bricks() looks for the volfiles and sends an rpc
 2949     request to glusterfs_handle_attach(). Now, since the volinfo
 2950     has been deleted by glusterd_delete_stale_volume()
 2951     from priv->volumes list before glusterd_start_bricks() and
 2952     glusterd_create_volfiles_and_notify_services() and
 2953     glusterd_list_add_order is called after glusterd_start_bricks(),
 2954     therefore the attach RPC req gets an empty volfile path
 2955     and that causes the brick to crash.
 2956     
 2957     Fix- Call glusterd_list_add_order() and
 2958     glusterd_create_volfiles_and_notify_services before
 2959     glusterd_start_bricks() cal is made in glusterd_import_friend_volume
 2960     
 2961     Change-Id: Idfe0e8710f7eb77ca3ddfa1cabeb45b2987f41aa
 2962     Fixes: bz#1773856
 2963     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
 2964 
 2965 commit 672a53d3d30d6932e25d81f64c7fecbdb97d15c4
 2966 Author: Xavi Hernandez <xhernandez@redhat.com>
 2967 Date:   Wed Feb 19 12:24:15 2020 +0100
 2968 
 2969     core: Prevent crash on process termination
 2970     
 2971     A previous patch (ce61da816a) has fixed a use-after-free issue,
 2972     but it doesn't work well when the final cleanup is done at process
 2973     termination because gluster doesn't stop other threads before
 2974     calling exit().
 2975     
 2976     For this reason, the final cleanup is removed to avoid the crash,
 2977     at least until the termination sequence properly stops all gluster
 2978     threads before exiting the program.
 2979     
 2980     Change-Id: Id7cfb4407fcf208e28f03a7c3cdc3ef9c1f3bf9b
 2981     Fixes: bz#1801684
 2982     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
 2983 
 2984 commit b1a18ef5d0ad7317f4c22cd9c765c63634100906
 2985 Author: Kaleb S. KEITHLEY <kkeithle@redhat.com>
 2986 Date:   Wed Feb 12 12:47:57 2020 -0500
 2987 
 2988     ganesha-ha: updates for pcs-0.10.x (i.e. in Fedora-29 and RHEL-8)
 2989     
 2990     pcs-0.10 has introduced changes options to pcs commands
 2991     
 2992     pcs-0.10.x is in Fedora-29 and later and RHEL-8.
 2993     
 2994     Also some minor cleanup. Namely use bash built-in [[...]] in a few
 2995     more places instead of test(1), i.e. [...], and use correct "==" for
 2996     comparison.
 2997     
 2998     Change-Id: I3fb2fcd71406964c77fdc4f18580ca133f365fd6
 2999     Fixes: bz#1193929
 3000     Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com>
 3001 
 3002 commit ee415f38f71bad62368333184fab3b70be524b78
 3003 Author: Xavi Hernandez <xhernandez@redhat.com>
 3004 Date:   Fri Feb 7 10:19:57 2020 +0100
 3005 
 3006     core: fix memory pool management races
 3007     
 3008     Objects allocated from a per-thread memory pool keep a reference to it
 3009     to be able to return the object to the pool when not used anymore. The
 3010     object holding this reference can have a long life cycle that could
 3011     survive a glfs_fini() call.
 3012     
 3013     This means that it's unsafe to destroy memory pools from glfs_fini().
 3014     
 3015     Another side effect of destroying memory pools from glfs_fini() is that
 3016     the TLS variable that points to one of those pools cannot be reset for
 3017     all alive threads.  This means that any attempt to allocate memory from
 3018     those threads will access already free'd memory, which is very
 3019     dangerous.
 3020     
 3021     To fix these issues, mem_pools_fini() doesn't destroy pool lists
 3022     anymore. Only at process termination the pools are destroyed.
 3023     
 3024     Change-Id: Ib189a5510ab6bdac78983c6c65a022e9634b0965
 3025     Fixes: bz#1801684
 3026     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
 3027 
 3028 commit f389b960d4d337e448808728bfaf2aa168cebdac
 3029 Author: Rishubh Jain <risjain@redhat.com>
 3030 Date:   Sun Aug 18 18:02:57 2019 +0530
 3031 
 3032     Updating gluster manual.
 3033     
 3034     Adding disperse-data to gluster manual under
 3035     volume create command
 3036     
 3037     Change-Id: Ic9eb47c9e71a1d7a11af9394c615c8e90f8d1d69
 3038     Fixes: bz#1668239
 3039     Signed-off-by: Rishubh Jain <risjain@redhat.com>
 3040     Signed-off-by: Sheetal Pamecha <spamecha@redhat.com>
 3041 
 3042 commit 06453d77d056fbaa393a137ca277a20e38d2f67e
 3043 Author: Ravishankar N <ravishankar@redhat.com>
 3044 Date:   Tue Feb 11 14:34:48 2020 +0530
 3045 
 3046     afr: prevent spurious entry heals leading to gfid split-brain
 3047     
 3048     Problem:
 3049     In a hyperconverged setup with granular-entry-heal enabled, if a file is
 3050     recreated while one of the bricks is down, and an index heal is triggered
 3051     (with the brick still down), entry-self heal was doing a spurious heal
 3052     with just the 2 good bricks. It was doing a post-op leading to removal
 3053     of the filename from .glusterfs/indices/entry-changes as well as
 3054     erroneous setting of afr xattrs on the parent. When the brick came up,
 3055     the xattrs were cleared, resulting in the renamed file not getting
 3056     healed and leading to gfid split-brain and EIO on the mount.
 3057     
 3058     Fix:
 3059     Proceed with entry heal only when shd can connect to all bricks of the replica,
 3060     just like in data and metadata heal.
 3061     
 3062     fixes: bz#1801624
 3063     Change-Id: I916ae26ad1fabf259bc6362da52d433b7223b17e
 3064     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
 3065 
 3066 commit a7fa54ddea3fe429f143b37e4de06a93b49d776a
 3067 Author: Ashish Pandey <aspandey@redhat.com>
 3068 Date:   Fri Jan 3 16:54:33 2020 +0530
 3069 
 3070     cluster/thin-arbiter: Wait for TA connection before ta-file lookup
 3071     
 3072     Problem:
 3073     When we mount a ta volume, as soon as 2 data bricks are connected
 3074     we consider that the mount is done and then send a lookup/create on
 3075     ta file on ta node. However, this connection with ta node might not
 3076     have been completed.
 3077     Due to this delay, ta replica id file will not be created and we
 3078     will see ENOTCONN error in log file if we do lookup.
 3079     
 3080     Solution:
 3081     As we know that this ta node could have a higher latency, we should
 3082     wait for reasonable time for connection to happen before sending
 3083     lookup/create on replica id file.
 3084     
 3085     fixes: bz#1720463
 3086     Change-Id: I36f90865afe617e4e84cee57fec832a16f5dd6cc
 3087 
 3088 commit acbba8e1121205f9512833d1b0fad81b0930f881
 3089 Author: yatipadia <ypadia@redhat.com>
 3090 Date:   Tue Nov 19 18:59:28 2019 +0530
 3091 
 3092     libglusterfs-xlator: structure logging
 3093     
 3094     convert all gf_msg() to gf_smsg()
 3095     
 3096     Change-Id: Id542e05faadb8041b472a2298c71fe62730e65fc
 3097     Updates: #657
 3098     Signed-off-by: yatipadia <ypadia@redhat.com>
 3099 
 3100 commit c048df835e94a8707254d32db93c42f43f31abda
 3101 Author: Barak Sason Rofman <bsasonro@redhat.com>
 3102 Date:   Wed Dec 18 19:18:40 2019 +0200
 3103 
 3104     dht - Reducing methods scope
 3105     
 3106     1. Reduced methods scope in the following:
 3107     inode read&write, layout, linkfile, shard
 3108     2. Removed dead code @ dht-linkkile.c:174-228 & dht-shard.c:44
 3109     
 3110     Change-Id: I2d08a10c7b074fccdb0c020845cad60c6ea32db5
 3111     updates: bz#1193929
 3112     Signed-off-by: Barak Sason Rofman <bsasonro@redhat.com>
 3113 
 3114 commit a3c8a6180a53fb96cc090c5552c64088f68ec4dc
 3115 Author: karthik-us <ksubrahm@redhat.com>
 3116 Date:   Wed Mar 13 13:00:47 2019 +0530
 3117 
 3118     cluster/afr: Check for lock on source & sink before doing data heal
 3119     
 3120     Problem:
 3121     In function afr_selfheal_data_block(), we only check for the lock count
 3122     to be equal to or greater than the number of sinks. There can be a case
 3123     where we have 2 source bricks and one sink and the locking is successful
 3124     on only the source brick(s). In this case we continue with the healing
 3125     on sink without having a lock, which is not correct.
 3126     
 3127     Fix:
 3128     Check for lock on atleast source & one sink before starting the data heal.
 3129     
 3130     Change-Id: Iebcb57dcaa4b31831fedfee63d6ca16e9d6c8df8
 3131     fixes: bz#1688115
 3132     Signed-off-by: karthik-us <ksubrahm@redhat.com>
 3133 
 3134 commit 8db8202f716fd24c8c52f8ee5f66e169310dc9b1
 3135 Author: Amar Tumballi <amar@kadalu.io>
 3136 Date:   Tue Feb 4 23:02:51 2020 +0530
 3137 
 3138     volgen: make thin-arbiter name unique in 'pending-xattr' option
 3139     
 3140     Thin-arbiter module makes use of 'pending-xattr' name for the translator
 3141     as the filename which gets created in thin-arbiter node. By making this
 3142     unique, we can host single thin-arbiter node for multiple clusters.
 3143     
 3144     Updates: #763
 3145     Change-Id: Ib3c732e7e04e6dba229e71ae3e64f1f3cb6d794d
 3146     Signed-off-by: Amar Tumballi <amar@kadalu.io>
 3147 
 3148 commit d27df94016b5526c18ee964d4a47508326329dda
 3149 Author: Pranith Kumar K <pkarampu@redhat.com>
 3150 Date:   Tue Feb 11 15:42:05 2020 +0530
 3151 
 3152     tests: Fix spurious self-heald.t failure
 3153     
 3154     Problem:
 3155     heal-info code assumes that all indices in xattrop directory
 3156     definitely need heal. There is one corner case.
 3157     The very first xattrop on the file will lead to adding the
 3158     gfid to 'xattrop' index in fop path and in _cbk path it is
 3159     removed because the fop is zero-xattr xattrop in success case.
 3160     These gfids could be read by heal-info and shown as needing heal.
 3161     
 3162     Fix:
 3163     Check the pending flag to see if the file definitely needs or
 3164     not instead of which index is being crawled at the moment.
 3165     
 3166     fixes: bz#1801623
 3167     Change-Id: I79f00dc7366fedbbb25ec4bec838dba3b34c7ad5
 3168     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 3169 
 3170 commit 6de80bcd6366778ac34ce58ec496fa08cc02bd0b
 3171 Author: Mohit Agrawal <moagrawal@redhat.com>
 3172 Date:   Sat Sep 29 13:15:35 2018 +0530
 3173 
 3174     feature/changelog: Avoid thread creation if xlator is not enabled
 3175     
 3176     Problem:
 3177     Changelog creates threads even if the changelog is not enabled
 3178     
 3179     Background:
 3180     Changelog xlator broadly does two things
 3181       1. Journalling - Cosumers are geo-rep and glusterfind
 3182       2. Event Notification for registered events like (open, release etc) -
 3183          Consumers are bitrot, geo-rep
 3184     
 3185     The existing option "changelog.changelog" controls journalling and
 3186     there is no option to control event notification and is enabled by
 3187     default. So when bitrot/geo-rep is not enabled on the volume, threads
 3188     and resources(rpc and rbuf) related to event notifications consumes
 3189     resources and cpu cycle which is unnecessary.
 3190     
 3191     Solution:
 3192     The solution is to have two different options as below.
 3193      1. changelog-notification : Event notifications
 3194      2. changelog : Journalling
 3195     
 3196     This patch introduces the option "changelog-notification" which is
 3197     not exposed to user. When either bitrot or changelog (journalling)
 3198     is enabled, it internally enbales 'changelog-notification'. But
 3199     once the 'changelog-notification' is enabled, it will not be disabled
 3200     for the life time of the brick process even after bitrot and changelog
 3201     is disabled. As of now, rpc resource cleanup has lot of races and is
 3202     difficult to cleanup cleanly. If allowed, it leads to memory leaks
 3203     and crashes on enable/disable of bitrot or changelog (journal) in a
 3204     loop. Hence to be safer, the event notification is not disabled within
 3205     lifetime of process once enabled.
 3206     
 3207     Change-Id: Ifd00286e0966049e8eb9f21567fe407cf11bb02a
 3208     Updates: #475
 3209     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 3210 
 3211 commit f1e62e1578ee07802827121722d90c1ec4b811a6
 3212 Author: yatipadia <ypadia@redhat.com>
 3213 Date:   Sat Nov 23 14:03:41 2019 +0530
 3214 
 3215     libglusterfs-options: structure logging
 3216     
 3217     convert all gf_msg() to gf_smsg()
 3218     
 3219     Change-Id: I8f1ff462b9c8012ed676c51450930a65ac403bf3
 3220     Updates: #657
 3221     Signed-off-by: yatipadia <ypadia@redhat.com>
 3222 
 3223 commit c87817495b3c5c36dcca9d157e9313b7d3195eed
 3224 Author: Susant Palai <spalai@redhat.com>
 3225 Date:   Fri Dec 27 12:06:19 2019 +0530
 3226 
 3227     dht: Fix stale-layout and create issue
 3228     
 3229     Problem: With lookup-optimize set to on by default, a client with
 3230     stale-layout can create a new file on a wrong subvol. This will lead to
 3231     possible duplicate files if two different clients attempt to create the
 3232     same file with two different layouts.
 3233     
 3234     Solution: Send in-memory layout to be cross checked at posix before
 3235     commiting a "create". In case of a mismatch, sync the client layout with
 3236     that of the server and attempt the create fop one more time.
 3237     
 3238     test: Manual, testcase(attached)
 3239     
 3240     fixes: bz#1786679
 3241     Change-Id: Ife0941f105113f1c572f4363cbcee65e0dd9bd6a
 3242     Signed-off-by: Susant Palai <spalai@redhat.com>
 3243 
 3244 commit 8fad76650bd85463708f59d2518f5b764ae4c702
 3245 Author: Kotresh HR <khiremat@redhat.com>
 3246 Date:   Mon Feb 3 18:10:17 2020 +0530
 3247 
 3248     bitrot: Make number of signer threads configurable
 3249     
 3250     The number of signing process threads (glfs_brpobj)
 3251     is set to 4 by default. The recommendation is to set
 3252     it to number of cores available. This patch makes it
 3253     configurable as follows
 3254     
 3255     gluster vol bitrot <volname> signer-threads <count>
 3256     
 3257     fixes: bz#1797869
 3258     Change-Id: Ia883b3e5e34e0bc8d095243508d320c9c9c58adc
 3259     Signed-off-by: Kotresh HR <khiremat@redhat.com>
 3260 
 3261 commit 2dbbdc92be205d12901ce2770570bbcb17040838
 3262 Author: yatipadia <ypadia@redhat.com>
 3263 Date:   Fri Nov 22 19:00:00 2019 +0530
 3264 
 3265     libglusterfs-event-epoll: structure logging
 3266     
 3267     convert gf_msg() to gf_smsg()
 3268     
 3269     Change-Id: Idf5bfc826b0c9f1a2674eea2a2e6164f30806b00
 3270     Updates: #657
 3271     Signed-off-by: yatipadia <ypadia@redhat.com>
 3272 
 3273 commit 228d27d7bd731aa37a26fe637ebcde4322463bb5
 3274 Author: yatipadia <ypadia@redhat.com>
 3275 Date:   Tue Nov 5 21:44:54 2019 +0530
 3276 
 3277     glusterfsd: structure logging
 3278     
 3279     convert gf_msg() to gf_smsg()
 3280     
 3281     Change-Id: I1cd6a5ac6f4361195d5d925efb2cc194045d0bba
 3282     Updates: #657
 3283     Signed-off-by: yatip <ypadia@redhat.com>
 3284 
 3285 commit 5864b450b62da45487050b4cd8215343f9b73606
 3286 Author: yatip <ypadia@redhat.com>
 3287 Date:   Thu Jan 30 18:59:54 2020 +0530
 3288 
 3289     protocol/client-rpc-fops: structure logging
 3290     
 3291     convert all gf_msg() to gf_smsg()
 3292     
 3293     Updates: #657
 3294     
 3295     Change-Id: Ia9d4fb17579af6586bc13d69ec7990c6cf220aac
 3296     Signed-off-by: yatip <ypadia@redhat.com>
 3297 
 3298 commit f121b156cca43148595229e07a9c5a5a4561627c
 3299 Author: Amar Tumballi <amarts@redhat.com>
 3300 Date:   Thu Aug 1 15:44:15 2019 +0530
 3301 
 3302     configure: add an option to build with tcmalloc library
 3303     
 3304     With this patch, one can now use '--enable-tcmalloc' while running
 3305     configure, and they can see that their glusterfs is linked with
 3306     libtcmalloc.
 3307     
 3308     [atumball@local build]$ ldd /usr/local/sbin/glusterfs | grep tcmalloc
 3309     	libtcmalloc.so.4 => /lib64/libtcmalloc.so.4 (0x00007feec0b87000)
 3310     
 3311     Once we establish a standard performance number with and without this option,
 3312     we will see how to make it default.
 3313     
 3314     Updates: #237
 3315     Change-Id: I3377f57bfe4e17f101a212e1914a6d3c1687d528
 3316     Signed-off-by: Amar Tumballi <amarts@redhat.com>
 3317 
 3318 commit 4e9fc11b0659957e76565f6557ad1079c51d6832
 3319 Author: Pranith Kumar K <pkarampu@redhat.com>
 3320 Date:   Tue Feb 4 14:43:35 2020 +0530
 3321 
 3322     protocol/client: Maintain connection to brick status
 3323     
 3324     Problem:
 3325     The current version of the code depends on the non-zero-port information to
 3326     propagate ping event from brick to parent layers. But as and when connection
 3327     succeeds, port is set to zero in rpc layer.So ping event is never propagated to
 3328     parent layers. Halo doesn't work without this.
 3329     
 3330     Fix:
 3331     Remember the status of connection in 'private' structure and use that to decide
 3332     to propagate ping event to parent xlator.
 3333     
 3334     fixes: bz#1797934
 3335     Change-Id: Ia578ba9fb3813953d2068dbba5c982ab27cc3429
 3336     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 3337 
 3338 commit 2b35572e1f30c37bbbbec8f63060ae2eeaf61b32
 3339 Author: Yaniv Kaul <ykaul@redhat.com>
 3340 Date:   Tue Jan 21 10:16:29 2020 +0200
 3341 
 3342     dht-hashfn.c: ensure we do not try to calculate hash on NULL path
 3343     
 3344     For some reason, dht_selfheal_layout_alloc_start() sends a NULL
 3345     loc->path string to dht_hash_compute().
 3346     Until we understand why it happens, we should strive not to crash
 3347     on a strlen of a NULL pointer.
 3348     
 3349     Change-Id: I8c2a22602cfccba9af85f432a1841556f6978450
 3350     updates: bz#1793378
 3351     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 3352 
 3353 commit c4b5cfbfa82de0aa16e9557f5b3fcf3dc186c740
 3354 Author: yatipadia <ypadia@redhat.com>
 3355 Date:   Tue Dec 10 17:10:44 2019 +0530
 3356 
 3357     xlator/dht-selfheal: structure logging
 3358     
 3359     Convert gf_msg() to gf_smsg()
 3360     
 3361     Change-Id: Ic72f2513e641cfcbe074933cb2697ee9fc05a766
 3362     Updates: #657
 3363     Signed-off-by: yatip <ypadia@redhat.com>
 3364 
 3365 commit ec2c7ddd6739afacc871a5df4c348b4c58cf153e
 3366 Author: yatipadia <ypadia@redhat.com>
 3367 Date:   Wed Dec 11 19:54:45 2019 +0530
 3368 
 3369     xlator/bit-rot: structure logging
 3370     
 3371     convert all gf_msg() to gf_smsg()
 3372     
 3373     Change-Id: Ifa45b5089f83ddfcf69132bb8d9c0dc6d012464b
 3374     Updates: #657
 3375     Signed-off-by: yatipadia <ypadia@redhat.com>
 3376 
 3377 commit ff69d2d52b88d87c6fba923ec9241f78528cee12
 3378 Author: yatip <ypadia@redhat.com>
 3379 Date:   Fri Jan 17 15:00:12 2020 +0530
 3380 
 3381     xlator/snapview: structure logging
 3382     
 3383     convert gf_msg() to gf_smsg()
 3384     
 3385     Updates: #657
 3386     
 3387     Change-Id: I2dd09a205c11d1c05abc7106e67802d2aa3bca79
 3388     Signed-off-by: yatip <ypadia@redhat.com>
 3389 
 3390 commit c0dce4603cb5dadc91245697f233736c41faec81
 3391 Author: yatipadia <ypadia@redhat.com>
 3392 Date:   Sun Nov 17 12:35:20 2019 +0530
 3393 
 3394     libglusterfs/common-utils: structure logging
 3395     
 3396     changes all the gf_msg() to gf_smgs()
 3397     
 3398     Change-Id: I3524bbd8f8070df2f2c888ea9b0fb7db1ee44453
 3399     Updates: #657
 3400     Signed-off-by: yatipadia <ypadia@redhat.com>
 3401 
 3402 commit 21684787fff47b25abf1ea06283bb4844114d429
 3403 Author: yatip <ypadia@redhat.com>
 3404 Date:   Thu Jan 16 12:54:43 2020 +0530
 3405 
 3406     xlator/dht-linkfile: structure logging
 3407     
 3408     convert all gf_msg() to gf_smsg()
 3409     
 3410     Updates: #657
 3411     
 3412     Change-Id: I9104ba8a8102f04d031a208abb06b6cf8ea8fd13
 3413     Signed-off-by: yatip <ypadia@redhat.com>
 3414 
 3415 commit 180302d33aab1522dbe8903f3b12c4ca95e474f4
 3416 Author: yatip <ypadia@redhat.com>
 3417 Date:   Wed Jan 29 18:02:05 2020 +0530
 3418 
 3419     protocol/client: structure logging
 3420     
 3421     convert all gf_msg() to gf_smsg()
 3422     
 3423     Updates: #657
 3424     
 3425     Change-Id: I69b228d7c7a8bc6263f9bd33710e880678d8c017
 3426     Signed-off-by: yatip <ypadia@redhat.com>
 3427 
 3428 commit 6df9dbf2edd7958981aa8bf9f0f8b3a9750bc635
 3429 Author: yatip <ypadia@redhat.com>
 3430 Date:   Fri Jan 24 20:58:06 2020 +0530
 3431 
 3432     protocol/client-handshake: structure logging
 3433     
 3434     convert gf_msg() to gf_smsg()
 3435     
 3436     Updates: #657
 3437     
 3438     Change-Id: Iee07228cfc3a9a9cd10e89ae9eb918681b072585
 3439     Signed-off-by: yatip <ypadia@redhat.com>
 3440 
 3441 commit 8b006854a58ea751f450f89de2b28b323c2c5c90
 3442 Author: yatip <ypadia@redhat.com>
 3443 Date:   Fri Jan 24 19:01:15 2020 +0530
 3444 
 3445     protocol/client-rpc-fops_v2: structure logging
 3446     
 3447     convert gf_msg() to gf_smsg()
 3448     
 3449     Updates: #657
 3450     
 3451     Change-Id: Ib45f121f583c2af09bfddb23391f73a117e63213
 3452     Signed-off-by: yatip <ypadia@redhat.com>
 3453 
 3454 commit a144cd20745a03f025920ab81265e150e11b3a0a
 3455 Author: Sheetal Pamecha <spamecha@redhat.com>
 3456 Date:   Sun Dec 22 22:52:30 2019 +0530
 3457 
 3458     ec: change error message for heal commands for disperse volume
 3459     
 3460     Currently when we issue a heal statistics or similar commands
 3461     for disperse volume, it fails with message "Volume is not of
 3462     type replicate." Adding message "this command is supported for
 3463     volumes of type replicate" to reflect supportability and better
 3464     understanding of heal functionality for disperse volumes.
 3465     
 3466     fixes: bz#1785998
 3467     Change-Id: I9688a9fdf427cb6f657cfd5b8db2f76a6c56f6e2
 3468     Signed-off-by: Sheetal Pamecha <spamecha@redhat.com>
 3469 
 3470 commit 34347876bb525e1af19bccafbb753a14b358d3aa
 3471 Author: Sunny Kumar <sunkumar@redhat.com>
 3472 Date:   Fri Jan 31 15:27:46 2020 +0000
 3473 
 3474     libgfapi: fix resource leak
 3475     
 3476     This patch fixes resource leak in glfs-fops.c.
 3477     CID : 1403873 : Resource leak
 3478     
 3479     updates: bz#789278
 3480     
 3481     Change-Id: I589d4d65d2b074801f7b4b0afd7dc4607d79420e
 3482     Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
 3483 
 3484 commit c1c2afd920ab6135ffcd0e1ecb25e71449037c9e
 3485 Author: Ashish Pandey <aspandey@redhat.com>
 3486 Date:   Wed Dec 4 17:06:18 2019 +0530
 3487 
 3488     Improve logging in EC, client and lock translator
 3489     
 3490     Change-Id: I98af8672a25ff9fd9dba91a2e1384719f9155255
 3491     Fixes: bz#1779760
 3492 
 3493 commit 06eb3d0b8488d12a92ea21b92727497c585bae4b
 3494 Author: Harpreet Kaur <hlalwani@redhat.com>
 3495 Date:   Mon Jan 7 16:38:25 2019 +0530
 3496 
 3497     geo-rep: Fix for "Transport End Point not connected" issue
 3498     
 3499     problem: Geo-rep gsyncd process mounts the master and slave volume
 3500              on master nodes and slave nodes respectively and starts
 3501              the sync. But it doesn't wait for the mount to be in ready
 3502              state to accept I/O. The gluster mount is considered to be
 3503              ready when all the distribute sub-volumes is up. If the all
 3504              the distribute subvolumes are not up, it can cause ENOTCONN
 3505              error, when lookup on file comes and file is on the subvol
 3506              that is down.
 3507     
 3508     solution: Added a Virtual Xattr "dht.subvol.status" which returns "1"
 3509               if all subvols are up and "0" if all subvols are not up.
 3510               Geo-rep then uses this virtual xattr after a fresh mount, to
 3511               check whether all subvols are up or not and then starts the
 3512               I/O.
 3513     
 3514     fixes: bz#1664335
 3515     Change-Id: If3ad01d728b1372da7c08ccbe75a45bdc1ab2a91
 3516     Signed-off-by: Harpreet Kaur <hlalwani@redhat.com>
 3517     Signed-off-by: Kotresh HR <khiremat@redhat.com>
 3518 
 3519 commit 12bea344e4cc4fe2101aa7efb8a97606d69aa80f
 3520 Author: Kaleb S. KEITHLEY <kkeithle@redhat.com>
 3521 Date:   Wed Jan 22 14:37:04 2020 -0500
 3522 
 3523     build: glusterfs-ganesha pkg requires python3-policycoreutils on rhel8
 3524     
 3525     glusterfs-ganesha pkg requires policycoreutils-python-utils on rhel8,
 3526     not policycoreutils-python
 3527     
 3528     Also needs nfs-ganesha-selinux on recent fedora and rhel/centos 8
 3529     
 3530     Change-Id: Ia97b4dabdc098fb76e3f60e8b48ea4191e677136
 3531     updates: bz#1193929
 3532     Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com>
 3533 
 3534 commit 7cac693bba33cbba35a6d0bbe98d8ad5551d2134
 3535 Author: Sunil Kumar Acharya <sheggodu@redhat.com>
 3536 Date:   Thu Nov 14 09:24:26 2019 +0530
 3537 
 3538     tests: fix test failures for nfsnobody user and group
 3539     
 3540     'nfsnobody' user and group is merged with
 3541     'nobody' user and group in RHEL8.
 3542     
 3543     Tests are modified to use appropriate user and group.
 3544     
 3545     BUG: 1756900
 3546     Change-Id: I59863da2262283b00b1cb417d3652ebe29a36407
 3547     Signed-off-by: Sunil Kumar Acharya <sheggodu@redhat.com>
 3548 
 3549 commit 7dc4122b84c93cfa6355002fa651a806706e4990
 3550 Author: Sheetal Pamecha <spamecha@redhat.com>
 3551 Date:   Thu Jan 2 12:05:12 2020 +0530
 3552 
 3553     afr: restore timestamp of files during metadata heal
 3554     
 3555     For files: During metadata heal, we restore timestamps
 3556     only for non-regular (char, block etc.) files.
 3557     Extenting it for regular files as timestamp is updated
 3558     via touch command also
 3559     
 3560     fixes: bz#1787274
 3561     Change-Id: I26fe4fb6dff679422ba4698a7f828bf62ca7ca18
 3562     Signed-off-by: Sheetal Pamecha <spamecha@redhat.com>
 3563 
 3564 commit 22b8564e781d2113327a0dc17dd4414a6386f2e1
 3565 Author: Xie Changlong <xiechanglong@cmss.chinamobile.com>
 3566 Date:   Sun Jan 19 17:54:51 2020 +0800
 3567 
 3568     snapview: fix memory leak
 3569     
 3570     Setting frame->local = local immediately after mem_get0.
 3571     In this case, SVC_STACK_UNWIND will take of freeing the memory.
 3572     
 3573     Change-Id: If6433be0e14ed0f4ab4127f70a8875f81bb3e6b8
 3574     updates: bz#789278
 3575     Signed-off-by: Xie Changlong <xiechanglong@cmss.chinamobile.com>
 3576 
 3577 commit 92bf7634d67a574dba6d7e77501d605d7f33acc3
 3578 Author: Yaniv Kaul <ykaul@redhat.com>
 3579 Date:   Tue Jan 14 17:11:15 2020 +0200
 3580 
 3581     server.c: fix Coverity issue 1405844 (memory - illegal access)
 3582     
 3583     We can use memcpy() instead of strncpy() as both are strings that are
 3584     37 bytes (GF_UUID_BUF_SIZE) long.
 3585     
 3586     fixes: CID#1405844
 3587     Change-Id: Ic74e8817cd790c13e29f3e6be8f18f2bfff77115
 3588     updates: bz#1193929
 3589     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 3590 
 3591 commit be673d9d093d977f4f38f2dba3f422c796db37b7
 3592 Author: Yaniv Kaul <ykaul@redhat.com>
 3593 Date:   Wed Jan 22 15:50:27 2020 +0200
 3594 
 3595     multiple files: ensure nfs_svc functions are available
 3596     
 3597     When gNFS is disabled in build time, we have to ensure
 3598     taht the .stop(), .start() and other functions of the nfs_svc
 3599     are not called, otherwise we'd crash.
 3600     In addition, #ifdef more code that is gNFS related.
 3601     
 3602     updates: bz#1793995
 3603     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 3604     
 3605     Change-Id: I5081f1670c04ca306aeaab7208829b0f2f149a42
 3606 
 3607 commit f6421dff22a6ddaf14134f6894deae219948c89d
 3608 Author: Mohit Agrawal <moagrawal@redhat.com>
 3609 Date:   Tue Jan 21 21:09:56 2020 +0530
 3610 
 3611     server: Mount fails after reboot 1/3 gluster nodes
 3612     
 3613     Problem: At the time of coming up one server node(1x3) after reboot
 3614     client is unmounted.The client is unmounted because a client
 3615     is getting AUTH_FAILED event and client call fini for the graph.The
 3616     client is getting AUTH_FAILED because brick is not attached with a
 3617     graph at that moment
 3618     
 3619     Solution: To avoid the unmounting the client graph throw ENOENT error
 3620               from server in case if brick is not attached with server at
 3621               the time of authenticate clients.
 3622     
 3623     Credits: Xavi Hernandez <xhernandez@redhat.com>
 3624     Change-Id: Ie6fbd73cbcf23a35d8db8841b3b6036e87682f5e
 3625     Fixes: bz#1793852
 3626     Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
 3627 
 3628 commit d8e685a96170a433dde93a6bc6fb8c07babb01a5
 3629 Author: Yaniv Kaul <ykaul@redhat.com>
 3630 Date:   Mon Oct 21 20:56:23 2019 +0300
 3631 
 3632     posix-entry-ops.c/posix.h: improve GFID_NULL_CHECK_AND_GOTO macro
 3633     
 3634     The macro already fetches from the dictionary the gfid-req variable.
 3635     Instead of throwing it away, keep it and in 2 cases,
 3636     re-use it later.
 3637     
 3638     updates: bz#1193929
 3639     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 3640     
 3641     Change-Id: Id6d42535132805ab69e7e3b802c140b90f2f8958
 3642 
 3643 commit 18549de12bcfafe4ac30fc2e11ad7a3f3c216b38
 3644 Author: Xi Jinyu <xijinyu@cmss.chinamobile.com>
 3645 Date:   Tue Jan 14 22:29:06 2020 -0500
 3646 
 3647     Fix possible resource leaks.
 3648     
 3649     xlators/features/quota/src/quota.c quota_log_usage function.
 3650     The quota_log_helper() function applies memory
 3651     for path through inode_path(), should be GF_FREE().
 3652     
 3653     fixes: bz#1792707
 3654     Change-Id: I33143bdf272bf10837061df4a1b7b2fc146162d5
 3655     Signed-off-by: Xi Jinyu <xijinyu@cmss.chinamobile.com>
 3656 
 3657 commit 61737898e84d30e02117baa7431f89cafff27f56
 3658 Author: Barak Sason Rofman <bsasonro@redhat.com>
 3659 Date:   Wed Dec 4 18:48:23 2019 +0200
 3660 
 3661     cli - removing fetch of unnecessary items.
 3662     
 3663     Somem methods dict_get...(...) values and then not use them anywhere.
 3664     Removed found occurrences.
 3665     
 3666     fixes: #753
 3667     
 3668     Change-Id: Iaeb8f4cec18f76078f6b2f4e4bd6f9795a3467bc
 3669     Signed-off-by: Barak Sason Rofman <bsasonro@redhat.com>
 3670 
 3671 commit 44602465081ea5fb38255ad68d3ed8e987190d1f
 3672 Author: Yaniv Kaul <ykaul@redhat.com>
 3673 Date:   Sun Jan 12 16:33:54 2020 +0200
 3674 
 3675     dictionary: remove the 'extra_free' parameter
 3676     
 3677     This parameter may have been used in the past, but is no longer
 3678     needed. Removing it and the few locations it was actually referenced.
 3679     This allows to remove an extra memdup as well, that was not needed
 3680     in the 1st place in server_setvolume() and unserialize_rsp_direntp()
 3681     functions.
 3682     
 3683     A followup separate patch will remove extra_stdfree parmeter
 3684     from the dictionary structure.
 3685     
 3686     Change-Id: Ica0ff0a330672373aaa60e808b7e76ec489a0fe3
 3687     updates: bz#1193929
 3688     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 3689 
 3690 commit 3aa0928623393aa9c296abf32d4726d02a454207
 3691 Author: Amar Tumballi <amar@kadalu.io>
 3692 Date:   Mon Jan 13 13:07:47 2020 +0530
 3693 
 3694     pkg-version: fix warning due to wrong check.
 3695     
 3696     using '==' operator is bash specific, and was giving
 3697     warnings like below when we ran `./autogen.sh`
 3698     
 3699     ```
 3700     build-aux/pkg-version: 26: [: x8dev: unexpected operator
 3701     build-aux/pkg-version: 49: [: x384.git22c2725bb: unexpected operator
 3702     ```
 3703     
 3704     This was not seen where 'sh' is linked to 'bash'.
 3705     
 3706     Updates: bz#1193929
 3707     Change-Id: Ied0d6707dd8b5d78756f302b6319b6ef75143018
 3708     Signed-off-by: Amar Tumballi <amar@kadalu.io>
 3709 
 3710 commit 3043fc54b0bdcf87fce13dcd9c56d0dd30e1c5b2
 3711 Author: Xie Changlong <xiechanglong@cmss.chinamobile.com>
 3712 Date:   Thu Jan 16 10:00:41 2020 +0800
 3713 
 3714     glusterd: fix use after free
 3715     
 3716     updates: bz#789278
 3717     Change-Id: I652d8d4428cf6ce61b712a66d309e78030a5f911
 3718     Signed-off-by: Xie Changlong <xiechanglong@cmss.chinamobile.com>
 3719 
 3720 commit c7b546326092d060bf84b7c578a4f21bc679674a
 3721 Author: Yaniv Kaul <ykaul@redhat.com>
 3722 Date:   Tue Jan 14 16:55:32 2020 +0200
 3723 
 3724     name.c: fix Coverity issues 1412332/3 - strcat into uninitialized value
 3725     
 3726     Check limit to 108 bytes before strcpy().
 3727     
 3728     fixes: CID#1412332
 3729     updates: bz#1193929
 3730     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 3731     
 3732     Change-Id: I8b26b1e1d2daca98ff36db531539bec0a405769c
 3733 
 3734 commit 485212e858bddd97573a3b2b811357b0d822005a
 3735 Author: Sunny Kumar <sunkumar@redhat.com>
 3736 Date:   Fri Jan 17 11:03:46 2020 +0000
 3737 
 3738     geo-rep: Fix ssh-port validation
 3739     
 3740     If non-standard ssh-port is used, Geo-rep can be configured to use ssh port
 3741     by using config option, the value should be in allowed port range and non negative.
 3742     
 3743     At present it can accept negative value and outside allowed port range which is incorrect.
 3744     
 3745     Many Linux kernels use the port range 32768 to 61000.
 3746     IANA suggests it should be in the range 1 to 2^16 - 1, so keeping the same.
 3747     
 3748     $ gluster volume geo-replication master 127.0.0.1::slave config ssh-port -22
 3749     geo-replication config updated successfully
 3750     $ gluster volume geo-replication master 127.0.0.1::slave config ssh-port 22222222
 3751     geo-replication config updated successfully
 3752     
 3753     This patch fixes the above issue and have added few validations around this
 3754     in test cases.
 3755     
 3756     Change-Id: I9875ab3f00d7257370fbac6f5ed4356d2fed3f3c
 3757     Fixes: bz#1792276
 3758     Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
 3759 
 3760 commit 6c48f1649d0916ae237fec08f007c8efb6d53e4e
 3761 Author: Xavi Hernandez <xhernandez@redhat.com>
 3762 Date:   Tue Jan 14 13:28:47 2020 +0100
 3763 
 3764     events: fix IPv6 memory corruption
 3765     
 3766     When an event was generated and the target host was resolved to an IPv6
 3767     address, there was a memory overflow when that address was copied to a
 3768     fixed IPv4 structure (IPv6 addresses are longer than IPv4 ones).
 3769     
 3770     This fix correctly handles IPv4 and IPv6 addresses returned by
 3771     getaddrinfo()
 3772     
 3773     Change-Id: I5864a0c6e6f1b405bd85988529570140cf23b250
 3774     Fixes: bz#1790870
 3775     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
 3776 
 3777 commit 71231a6a009ea34ea80da9b36d761f3764043de5
 3778 Author: Michael Scherer <misc@fedoraproject.org>
 3779 Date:   Thu Jan 16 11:21:09 2020 +0100
 3780 
 3781     Be explicit on this being a python3 script
 3782     
 3783     While the script seems to work on both python3 and python2, this
 3784     break the build of rawhide RPM who requires script to be either
 3785     using python2 or python3.
 3786     
 3787     Since python2 is going to be deprecated, I guess we should aim for
 3788     python3.
 3789     
 3790     Change-Id: Ic6322ad47772d708b60b96652a1122ee4a54141d
 3791     Fixes: bz#1791682
 3792     Signed-off-by: Michael Scherer <misc@fedoraproject.org>
 3793 
 3794 commit 43d58660772da0b8f25fb89ece19851391bf999f
 3795 Author: Anoop C S <anoopcs@redhat.com>
 3796 Date:   Tue Dec 10 11:43:30 2019 +0530
 3797 
 3798     protocol/client: Fix a log entry
 3799     
 3800     It has been a while since we removed lock healing logic from protocol
 3801     client. So no need to mention that we healed locks after fd reopen.
 3802     
 3803     Change-Id: I24bd3f9e9f2942e306714b2cb83c229ae57c60ae
 3804     Fixes: bz#1193929
 3805     Signed-off-by: Anoop C S <anoopcs@redhat.com>
 3806 
 3807 commit d5fef2a50cf4e3458e800d59021a97c0b5fc39db
 3808 Author: Yaniv Kaul <ykaul@redhat.com>
 3809 Date:   Mon Dec 9 23:17:16 2019 +0200
 3810 
 3811     dht-hashfn.c: remove a strlen()
 3812     
 3813     We already have the length of the name, or when we munge it,
 3814     we can return the length of it instead of strlen() again.
 3815     
 3816     Also, reduce a bit the code under the lock.
 3817     
 3818     Change-Id: I0141b0725ed1a4134d8d9f81ed1187b551b038b5
 3819     updates: bz#1193929
 3820     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 3821 
 3822 commit 9969d1dc2a3e815b161ce8a3dc5d08f84cfe011f
 3823 Author: Yaniv Kaul <ykaul@redhat.com>
 3824 Date:   Mon Dec 9 21:28:00 2019 +0200
 3825 
 3826     multiple xlators: reduce key length
 3827     
 3828     In many cases, we were freely allocating long keys with no need.
 3829     Smaller char arrays are just fine almost anywhere, so just went ahead
 3830     and looked where they we can use smaller ones.
 3831     
 3832     In some cases, annotated the functions as static and the prefixes
 3833     passed as const as it was easier to read and understand.
 3834     
 3835     Where relevant, converted the dict functions to use known key length.
 3836     
 3837     Change-Id: I882ab33ea20d90b63278336cd1370c09ffdab7f2
 3838     updates: bz#1193929
 3839     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 3840 
 3841 commit dc5b2d3c181020da1644106e5b17f77a4f3e53df
 3842 Author: yatipadia <ypadia@redhat.com>
 3843 Date:   Wed Dec 4 12:09:12 2019 +0530
 3844 
 3845     xlator/changelog: structure logging
 3846     
 3847     convert all gf_msg() to gf_smsg()
 3848     
 3849     Change-Id: I2303aeaa0775b663e2ea5741166a457cd747d2ac
 3850     Updates: #657
 3851     Signed-off-by: yatipadia <ypadia@redhat.com>
 3852     Signed-off-by: yatip <ypadia@redhat.com>
 3853 
 3854 commit 830a3dc46438b1cf18bcd35cca05c164e77180f1
 3855 Author: yatipadia <ypadia@redhat.com>
 3856 Date:   Mon Nov 25 19:57:09 2019 +0530
 3857 
 3858     api-glfs-fops: structure logging
 3859     
 3860     Convert gf_msg() to gf_smsg()
 3861     
 3862     Change-Id: Iceb40d60a75106c7b4b0a2487b611f7376bccc46
 3863     Updates: #657
 3864     Signed-off-by: yatipadia <ypadia@redhat.com>
 3865 
 3866 commit d73872e764214f8071c8915536a75bdac1e5e685
 3867 Author: Shwetha K Acharya <sacharya@redhat.com>
 3868 Date:   Tue Jan 14 10:51:06 2020 +0530
 3869 
 3870     tools/glusterfind: Remove an extra argument
 3871     
 3872     fixes: bz#1790748
 3873     Change-Id: I1cb12c975142794139456d0f8e99fbdbb03c53a1
 3874     Signed-off-by: Shwetha K Acharya <sacharya@redhat.com>
 3875 
 3876 commit 6b5068f3404c072b94e18854aab8e99d3ce54350
 3877 Author: yatipadia <ypadia@redhat.com>
 3878 Date:   Thu Dec 12 16:17:22 2019 +0530
 3879 
 3880     xlator/bit-rot-stub-helpers: structure logging
 3881     
 3882     convert gf_msg() to gf_smsg()
 3883     
 3884     Change-Id: Id03b6a2f6817101650ea6eeffb10e91c09e6ef80
 3885     Updates: #657
 3886     Signed-off-by: yatipadia <ypadia@redhat.com>
 3887 
 3888 commit 8bfcae18c080c856a7218b0df9e403e2cdb59355
 3889 Author: Dmitry Antipov <dmantipov@yandex.ru>
 3890 Date:   Fri Jan 10 13:44:53 2020 +0300
 3891 
 3892     xlators/storage: remove duplicated includes
 3893     
 3894     Do not include ftw.h twice.
 3895     
 3896     Change-Id: Id9e8d1813aafd890940adcd6883d90fa1b4beaf9
 3897     Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru>
 3898     Updates: bz#1193929
 3899 
 3900 commit 7dcbf096087cb6b939149c6f14ea2c3b2ab861e4
 3901 Author: Dmitry Antipov <dmantipov@yandex.ru>
 3902 Date:   Thu Jan 9 16:17:57 2020 +0300
 3903 
 3904     xlators/storage: prefer HAVE_xxx to host C library predefines
 3905     
 3906     Rely on configure-time HAVE_SET_FSID instead of predefined
 3907     linux and __GLIBC__, in the same way as fusermount.c does.
 3908     
 3909     Change-Id: If81f821014bbf3fdc43950670e316b4d178b4cac
 3910     Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru>
 3911     Updates: bz#1193929
 3912 
 3913 commit a73baf4e50fd8468f726179d89ba90364c5b8329
 3914 Author: Yaniv Kaul <ykaul@redhat.com>
 3915 Date:   Wed Oct 23 22:38:59 2019 +0300
 3916 
 3917     dict: use the free_pair's key presence or NULL as a sign of use.
 3918     
 3919     Instead of using a boolean parameter, we can use the key variable.
 3920     If it's NULL, the pair is not used and can be used.
 3921     Otherwise, it's in use - don't use.
 3922     
 3923     It saves this annoying boolean, which causes the compiler
 3924     (or us explicitly) to pad with additional bytes the dict struct.
 3925     
 3926     Change-Id: I89f52db57f35b3ef8acf57b7de2cee37f5d18e06
 3927     updates: bz#1193929
 3928     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 3929 
 3930 commit 905db496aa622e421d807222dcc488488da9dbfe
 3931 Author: Yaniv Kaul <ykaul@redhat.com>
 3932 Date:   Tue Dec 31 09:11:08 2019 +0200
 3933 
 3934     socket.c/name.c: minor changes
 3935     
 3936     - Move functions to static
 3937     - Remove redundant checks
 3938     - Use dict_get_...sizen() where applicable
 3939     - Remove unused variables.
 3940     - Moved some code to be executed only if relevant.
 3941     
 3942     ~3% object size reduction.
 3943     
 3944     Change-Id: Id9b8414e0a17442f1dac10ba77014d565756c935
 3945     updates: bz#1193929
 3946     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 3947 
 3948 commit c73adc852a6f94844bab9cf541838934dd8d45d4
 3949 Author: Yaniv Kaul <ykaul@redhat.com>
 3950 Date:   Tue Dec 31 15:11:42 2019 +0200
 3951 
 3952     rpc-transport: minor changes
 3953     
 3954     - Removed dead code
 3955     - Remove redundant checks
 3956     - Changed dict functions to use dict_..._sizen() functions.
 3957     
 3958     Change-Id: If00aaa90eef4078effd5b7fed2294f872e001b0a
 3959     updates: bz#1193929
 3960     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 3961 
 3962 commit 9189ed3201f1da89f16595d685b67a71e075d5be
 3963 Author: Yaniv Kaul <ykaul@redhat.com>
 3964 Date:   Fri Jan 10 08:24:35 2020 +0200
 3965 
 3966     glusterd-volgen.c: fix to disable property when gNFS disabled.
 3967     
 3968     The code was not ifdef'ed properly when gNFS was not enabled.
 3969     Strangely, I could not reproduce the failure on my system (Fedora 31),
 3970     but it was reproduced elsehwere and the fix was verified to correct it.
 3971     The failure:
 3972     gluster volume create testvol replica 3 127.0.0.2:/tests/brick{1..3} force
 3973     gluster v set testvol write-behind off
 3974     grep -rne write-behind /var/lib/glusterd/vols/testvol/trusted-testvol.tcp-fuse.vol
 3975     
 3976     The last grep was supposed to come out empty.
 3977     
 3978     The issue was that perfxl_option_handler may not have been called when it should
 3979     have been.
 3980     
 3981     Change-Id: Ie9f8ec87dabeef6624527c2266ddf9af01ca7373
 3982     updates: bz#1193929
 3983     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 3984 
 3985 commit 33c3cbe71b67f523538b04334f1ef962953281ed
 3986 Author: Sunny Kumar <sunkumar@redhat.com>
 3987 Date:   Fri Jan 10 14:28:35 2020 +0000
 3988 
 3989     glusterfind: python3 compatibility
 3990     
 3991     Problem:
 3992     While we delete gluster volume the hook script 'S57glusterfind-delete-post.py'
 3993     is failed to execute and error message can be observed in glusterd log.
 3994     
 3995     Traceback:
 3996       File "/var/lib/glusterd/hooks/1/delete/post/S57glusterfind-delete-post", line 69, in <module>
 3997         main()
 3998       File "/var/lib/glusterd/hooks/1/delete/post/S57glusterfind-delete-post", line 39, in main
 3999         glusterfind_dir = os.path.join(get_glusterd_workdir(), "glusterfind")
 4000       File "/usr/lib64/python3.7/posixpath.py", line 94, in join
 4001         genericpath._check_arg_types('join', a, *p)
 4002       File "/usr/lib64/python3.7/genericpath.py", line 155, in _check_arg_types
 4003         raise TypeError("Can't mix strings and bytes in path components") from None
 4004     TypeError: Can't mix strings and bytes in path components
 4005     
 4006     Solution:
 4007     
 4008     Added the 'universal_newlines' flag to Popen to support backward compatibility.
 4009     
 4010     Change-Id: Ie5655b11b55535c5ad2338108d0448e6fdaacf4f
 4011     Fixes: bz#1789478
 4012     Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
 4013 
 4014 commit 45894c39a4d05ed1f6a6f1bdbeafb5fe74ef29c3
 4015 Author: Kotresh HR <khiremat@redhat.com>
 4016 Date:   Fri Jan 10 16:48:14 2020 +0530
 4017 
 4018     glusterfind: Fix py2/py3 issues
 4019     
 4020     1. In dictionary values(), returns list in py2 and not in py3.
 4021        So explicitly convert it into list.
 4022     2. xattr module returns values in bytes. So explicitly convert
 4023        them to str to work both with py2 and py3
 4024     
 4025     fixes: bz#1789439
 4026     Change-Id: I27a639cda4f7a4ece9744a97c3d16e247906bd94
 4027     Signed-off-by: Kotresh HR <khiremat@redhat.com>
 4028 
 4029 commit 497d9f7d51443f4f463ca00f4fdddc42c6464e0f
 4030 Author: Yaniv Kaul <ykaul@redhat.com>
 4031 Date:   Tue Jan 7 12:57:29 2020 +0200
 4032 
 4033     dht-rename.c: fix Coverity issues 1397018/7 - strcat into uninitialized value
 4034     
 4035     initialize both src and dst if they were not initialized already.
 4036     
 4037     fixes: CID#1397018
 4038     Change-Id: Ic91954423953e8bf24eaa11fc2798c554f304d28
 4039     updates: bz#1193929
 4040     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 4041 
 4042 commit 00c090b093c147a95bfb8fce93f08303993e1995
 4043 Author: Xavi Hernandez <xhernandez@redhat.com>
 4044 Date:   Fri Dec 20 14:14:32 2019 +0100
 4045 
 4046     multiple: fix bad type cast
 4047     
 4048     When using inode_ctx_get() or inode_ctx_set(), a 'uint64_t *' is expected.
 4049     In many cases, the value to retrieve or store is a pointer, which will be
 4050     of smaller size in some architectures (for example 32-bits). In this case,
 4051     directly passing the address of the pointer casted to an 'uint64_t *' is
 4052     wrong and can cause memory corruption.
 4053     
 4054     Change-Id: Iae616da9dda528df6743fa2f65ae5cff5ad23258
 4055     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
 4056     Fixes: bz#1785611
 4057 
 4058 commit 1166df1920dd9b2bd5fce53ab49d27117db40238
 4059 Author: Csaba Henk <csaba@redhat.com>
 4060 Date:   Tue Jan 7 19:43:05 2020 +0100
 4061 
 4062     fuse: degrade logging of write failure to fuse device
 4063     
 4064     Problem:
 4065     
 4066     FUSE uses failures of communicating with /dev/fuse with various
 4067     errnos to indicate in-kernel conditions to userspace. Some of these
 4068     shouldn't be handled as an application error. Also the standard
 4069     POSIX errno description should not be shown as they are misleading
 4070     in this context.
 4071     
 4072     Solution:
 4073     
 4074     When writing to the fuse device, the caller of the respective
 4075     convenience routine can mask those errnos which don't qualify to
 4076     be an error for the application in that context, so then those
 4077     shall be reported at DEBUG level.
 4078     
 4079     The possible non-standard errnos are reported with their
 4080     POSIX name instead of their description to avoid confusion.
 4081     (Eg. for ENOENT we don't log "no such file or directory",
 4082     we log indeed literal "ENOENT".)
 4083     
 4084     Change-Id: I510158843e4b1d482bdc496c2e97b1860dc1ba93
 4085     updates: bz#1193929
 4086     Signed-off-by: Csaba Henk <csaba@redhat.com>
 4087 
 4088 commit 8d55951efe744f6da36f5b61c5be5a8bd77dc426
 4089 Author: Purna Pavan Chandra Aekkaladevi <paekkala@redhat.com>
 4090 Date:   Fri Oct 25 11:28:00 2019 +0530
 4091 
 4092     leases: re-alignment of structures
 4093     
 4094     All the structs present inside xlators/features/leases have been
 4095     re-aligned into memory efficient way. Changes are made in such a way
 4096     that every members are 8 byte aligned and no padding is left out in
 4097     the 8 byte line. Manual padding has been added to remove compile time
 4098     padding (which happens at the end of the struct). This manual padding
 4099     is for development and debugging benefits. It lets to use -Wpadded
 4100     option and raise warnings from the remaining structs where compile time
 4101     padding happens.
 4102     
 4103     Change-Id: I083d2de80e24f363c97adadcba64825e213050fc
 4104     Updates: bz#1754448
 4105     Signed-off-by: Purna Pavan Chandra Aekkaladevi <paekkala@redhat.com>
 4106 
 4107 commit b49f5e217fe958517464a84d0cc4b766d99fc42b
 4108 Author: Sanju Rakonde <srakonde@redhat.com>
 4109 Date:   Tue Jan 7 15:32:13 2020 +0530
 4110 
 4111     glusterd: increase the StartLimitBurst
 4112     
 4113     Based on https://bugzilla.redhat.com/show_bug.cgi?id=1782200#c6
 4114     increasing the limit.
 4115     
 4116     fixes: bz#1782200
 4117     
 4118     Change-Id: Ia885c7bdb2a90f0946c5268da894f6a4da5a69b7
 4119     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
 4120 
 4121 commit 144972bbddc4071600a1cc0380a997ada05c013d
 4122 Author: Ravishankar N <ravishankar@redhat.com>
 4123 Date:   Fri Jan 3 09:59:29 2020 +0530
 4124 
 4125     afr: expose cluster.optimistic-change-log to CLI.
 4126     
 4127     This volume option was not made avaialble to `gluster volume set` CLI.
 4128     
 4129     Reported-by: epolakis(https://github.com/kinsu) in
 4130     https://github.com/gluster/glusterfs/issues/781
 4131     
 4132     fixes: bz#1787554
 4133     Change-Id: I7141bdd4e53ee99e22b354edde8d023bfc0b2cd7
 4134     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
 4135 
 4136 commit 10b0d19513100c3f7edf00936ea5ec06fc3f80a5
 4137 Author: Barak Sason Rofman <bsasonro@redhat.com>
 4138 Date:   Sun Jan 5 12:11:07 2020 +0200
 4139 
 4140     cli - fixing a coverity issue
 4141     
 4142     Removed unused variable.
 4143     
 4144     fixes: CID#1412106
 4145     updates: bz#789278
 4146     
 4147     Change-Id: I1d4e1c1625cecf882d51e9cf4f5290383f63d405
 4148     Signed-off-by: Barak Sason Rofman <bsasonro@redhat.com>
 4149 
 4150 commit 50ca15e14e287f366576f257d715318d2514d7fa
 4151 Author: He Min <hemin@cmss.chinamobile.com>
 4152 Date:   Thu Jan 2 14:26:22 2020 +0800
 4153 
 4154     mgmt/glusterd: Correct the log path of valgrind.
 4155     
 4156     Problem: Currently enabling run-with-valgrind will cause the gnfs and quota to
 4157     fail to start. The phenomenon is as follows.
 4158     ------------------------------------------------------------------------------
 4159     NFS Server on localhost                     2049      0          N       48406
 4160     Quota Daemon on localhost                   N/A       N/A        N       48428
 4161     ------------------------------------------------------------------------------
 4162     
 4163     Solution: The cause of the above phenomenon is that the log path of valgrind is
 4164     set incorrectly. Gnfs and quota can start with valgrind normally after correcting
 4165     the log path.
 4166     
 4167     Updates: #788
 4168     Change-Id: Ib91408c08522ff66afff908fbff3fce4b93ea770
 4169     Signed-off-by: He Min <hemin@cmss.chinamobile.com>
 4170 
 4171 commit 26fe6106762f6aea910d477d1c37e645797884e1
 4172 Author: yatipadia <ypadia@redhat.com>
 4173 Date:   Tue Dec 17 12:55:02 2019 +0530
 4174 
 4175     xlator/performance/io-threads: structure logging
 4176     
 4177     convert gf_msg() to gf_smsg()
 4178     
 4179     Change-Id: I35c6f62c346a75ecb22cd3a4346ad4dc48f09a91
 4180     Updates: #657
 4181 
 4182 commit 7f98c105a3a8e2a9f4e9c75451a94115e270405e
 4183 Author: yatipadia <ypadia@redhat.com>
 4184 Date:   Wed Dec 18 17:25:51 2019 +0530
 4185 
 4186     xlator/io-cache: structure logging
 4187     
 4188     Convert all gf_msg() to gf_smsg()
 4189     
 4190     Updates: #657
 4191     
 4192     Change-Id: I72215b2518df78174dda8a7bc8de6f21fe1ba10f
 4193     Signed-off-by: yatipadia <ypadia@redhat.com>
 4194 
 4195 commit f72e2b00a13709717a6c7e0ff95d4c7a8f58b020
 4196 Author: Yaniv Kaul <ykaul@redhat.com>
 4197 Date:   Sun Dec 1 13:57:28 2019 +0200
 4198 
 4199     posix-metadata.c: try to getxattr in one call.
 4200     
 4201     Another location where instead of 2 sys calls we strive to get the
 4202     xattr in a single call, by guesstimating the required size
 4203     And avoid (or try to) not to first read the xattr len,
 4204     then another call to actually fetch. Instead, use a sane size
 4205     (256 bytes - worth checking if it makes sense or by default
 4206     use a larger size), and see if we can fetch it.
 4207     If we fail, we'll read the size and re-fetch.
 4208     
 4209     Such changes are needed elsewhere too (see
 4210     https://github.com/gluster/glusterfs/issues/720 )
 4211     
 4212     Change-Id: I466cea9d8b12fc45f6b37d202b1294ca28cd1fdd
 4213     updates: bz#1193929
 4214     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 4215 
 4216 commit f66bd85af09397300ad434655fc68861f48c2e3c
 4217 Author: Kaleb S. KEITHLEY <kkeithle@redhat.com>
 4218 Date:   Thu Jan 2 07:46:23 2020 -0500
 4219 
 4220     cli: duplicate defns of cli_default_conn_timeout and cli_ten_minutes_timeout
 4221     
 4222     Winter is coming. So is gcc-10.
 4223     
 4224     Compiling with gcc-10-20191219 snapshot reveals dupe defns of
 4225     cli_default_conn_timeout and cli_ten_minutes_timeout in
 4226     .../cli/src/cli.[ch] due to missing extern decl.
 4227     
 4228     There are many changes coming in gcc-10 described in
 4229     https://gcc.gnu.org/gcc-10/changes.html
 4230     
 4231     compiling cli.c with gcc-9 we see:
 4232        ...
 4233             .quad   .LC88
 4234             .comm   cli_ten_minutes_timeout,4,4
 4235             .comm   cli_default_conn_timeout,4,4
 4236             .text
 4237        .Letext0:
 4238        ...
 4239     
 4240     and with gcc-10:
 4241        ...
 4242             .quad   .LC88
 4243             .globl  cli_ten_minutes_timeout
 4244             .bss
 4245             .align 4
 4246             .type   cli_ten_minutes_timeout, @object
 4247             .size   cli_ten_minutes_timeout, 4
 4248        cli_ten_minutes_timeout:
 4249             .zero   4
 4250             .globl  cli_default_conn_timeout
 4251             .align 4
 4252             .type   cli_default_conn_timeout, @object
 4253             .size   cli_default_conn_timeout, 4
 4254        cli_default_conn_timeout:
 4255             .zero   4
 4256             .text
 4257        .Letext0:
 4258        ...
 4259     
 4260     which is reflected in the .o file as (gcc-9):
 4261     ...
 4262     0000000000000004 C cli_ten_minutes_timeout
 4263     0000000000000004 C cli_default_conn_timeout
 4264     ...
 4265     
 4266     and (gcc-10):
 4267     ...
 4268     0000000000000020 B cli_ten_minutes_timeout
 4269     0000000000000024 B cli_default_conn_timeout
 4270     ...
 4271     
 4272     See nm(1) and ld(1) for a description C (common) and B (BSS) and how
 4273     they are treated by the linker.
 4274     
 4275     Note: there is still a small chance that gcc-10 will land in Fedora-32,
 4276     despite 31 Dec. 2019 having been the deadline for that to happen.
 4277     
 4278     Change-Id: I54ea485736a4910254eeb21222ad263721cdef3c
 4279     Fixes: bz#1193929
 4280     Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com>
 4281 
 4282 commit 063bfe949eb11b33424bd67dd4225bb430ceaae7
 4283 Author: Xie Changlong <xiechanglong@cmss.chinamobile.com>
 4284 Date:   Mon Jan 6 10:11:32 2020 +0800
 4285 
 4286     lock: fix Coverity CID 1412107
 4287     
 4288     To avoid memory leak.
 4289     
 4290     Change-Id: Ib802be310fe989223afd623d3e5a9c9016d4fd38
 4291     updates: bz#789278
 4292     Signed-off-by: Xie Changlong <xiechanglong@cmss.chinamobile.com>
 4293 
 4294 commit f805eb110081783690d29ffe2d80405a9ba6191c
 4295 Author: kshithijiyer <kshithij.ki@gmail.com>
 4296 Date:   Tue Dec 24 13:02:21 2019 +0530
 4297 
 4298     [geo-rep] Improving help message in schedule_georep.py.in
 4299     
 4300     SLAVE positional argument doesn't provide a clear
 4301     picture of what it is when compared to mastervol and slavevol
 4302     in schedule_georep.py.in. It would be better if we change it to
 4303     something like "Slave hostame (<username>@SLAVEHOST or SLAVEHOST)"
 4304     
 4305     Present:
 4306     ----------
 4307     positional arguments:
 4308       mastervol            Master Volume Name
 4309       SLAVE                SLAVEHOST or root@SLAVEHOST or user@SLAVEHOST
 4310       slavevol             Slave Volume Name
 4311     
 4312     Suggested:
 4313     -----------
 4314     positional arguments:
 4315       mastervol            Master Volume Name
 4316       SLAVE                Slave hostname (<username>@SLAVEHOST or SLAVEHOST)
 4317       slavevol             Slave Volume Name
 4318     
 4319     fixes: bz#1786276
 4320     
 4321     Change-Id: I73d52247997d623f77d55e51cbb6eccc08eb95ff
 4322     Signed-off-by: kshithijiyer <kshithij.ki@gmail.com>
 4323 
 4324 commit 59ed165a60a58db62a67a9d9fbe75d8338e5ddeb
 4325 Author: Shwetha K Acharya <sacharya@redhat.com>
 4326 Date:   Wed Jul 31 11:34:19 2019 +0530
 4327 
 4328     geo-rep: Note section is required for ignore_deletes
 4329     
 4330     There exists a window of 15 sec, where the deletes are picked up
 4331     by history crawl when the ignore_deletes is set to true.
 4332     And it eventually deletes the file/s from slave which is/are not
 4333     supposed to be deleted. Though it is working as per design, a
 4334     note regarding this is needed.
 4335     
 4336     Added a warning message indicating the same.
 4337     Also logged info when the worker restarts after ignore-deletes
 4338     option set.
 4339     
 4340     fixes: bz#1708603
 4341     Change-Id: I103be882fac18b4cef935efa355f5037a396f7c1
 4342     Signed-off-by: Shwetha K Acharya <sacharya@redhat.com>
 4343 
 4344 commit 1e56e7acc3ced4f49d4494556bf6c86cd017b4ea
 4345 Author: Yaniv Kaul <ykaul@redhat.com>
 4346 Date:   Tue Dec 10 10:56:55 2019 +0200
 4347 
 4348     posix: improve posix_{set_gfid2path_xattr|remove_gfid2path_xattr} funcs
 4349     
 4350     1. Move them to posix-entry-ops.c and make them static.
 4351     2. Remove useless 'val' parameter allocation and snprintf() in
 4352     posix_set_gfid2path_xattr(), as it was identical to pgfid_bname.
 4353     
 4354     Change-Id: I41f3946c6c42aee9d7cf1150e21b442438fbcce2
 4355     updates: bz#1193929
 4356     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 4357 
 4358 commit 5cfd4553d163b96c9c409db68264d1b6af2fe49e
 4359 Author: Yaniv Kaul <ykaul@redhat.com>
 4360 Date:   Sun Dec 29 17:35:13 2019 +0200
 4361 
 4362     afr: simplify afr_has_quorum()
 4363     
 4364     1. Perform AFR_COUNT() once, in afr_has_quorum() and pass the result
 4365     to afr_lookup_has_quorum()
 4366     2. Simplify afr_lookup_has_quorum() - pass less parameters to it.
 4367     (Via the change in item 1 above).
 4368     3. Make afr_is_add_replica_mount_lookup_on_root() static function.
 4369     4. Remove dead code - afr_decide_heal_info() which was not used.
 4370     
 4371     Change-Id: If9168cd01e22788a0e60b91e315787d2aa60e97b
 4372     updates: bz#1193929
 4373     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 4374 
 4375 commit 51c4002030da6b1a11f6e8f48bf117084b1da34d
 4376 Author: Yaniv Kaul <ykaul@redhat.com>
 4377 Date:   Thu Dec 26 09:29:00 2019 +0200
 4378 
 4379     Remove some compiler warnings when not compiling gNFS
 4380     
 4381     As a follow up to https://review.gluster.org/#/c/glusterfs/+/23799/
 4382     When compiling without gNFS, there were some 'unused' warnings by
 4383     the compiler. This patch fixes them.
 4384     
 4385     Change-Id: I621562261f53950e821a450e0e7da304d00ae557
 4386     updates: bz#1193929
 4387     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 4388 
 4389 commit e40d0c399333e1ef62ff104b5cd35d8a5baa75f4
 4390 Author: Sanju Rakonde <srakonde@redhat.com>
 4391 Date:   Wed Dec 25 21:56:32 2019 +0530
 4392 
 4393     glusterd: deafult options after volume reset
 4394     
 4395     Problem: default option itransport.address-family is disappered
 4396     in volume info output after a volume reset.
 4397     
 4398     Cause: with 3.8.0 onwards volume option transport.address-family
 4399     has default value, any volume which is created will have this
 4400     option set. So, volume info will show this in its output. But,
 4401     with reset volume, this option is not handled.
 4402     
 4403     Solution: In glusterd_enable_default_options(), we should add this
 4404     option along with other default options. This function is called
 4405     by glusterd_options_reset() with volume reset command.
 4406     
 4407     fixes: bz#1786478
 4408     
 4409     Change-Id: I58f7aa24cf01f308c4efe6cae748cc3bc8b99b1d
 4410     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
 4411 
 4412 commit 7555ebc629774b95a878678f8ab0946c3bcf2f32
 4413 Author: Dmitry Antipov <dmantipov@yandex.ru>
 4414 Date:   Fri Dec 27 13:35:33 2019 +0300
 4415 
 4416     cli, libglusterfs: rely on libpthread static initializers
 4417     
 4418     cli-cmd.c: drop cli_cmd_cond_init() because static mutex and
 4419     condition variable are initialized with PTHREAD_xxx_INITIALIZERs.
 4420     
 4421     syncop-utils.c: since assignment PTHREAD_xxx_INITIALIZERs are
 4422     never failed, a few overengineered bits may be simplified.
 4423     
 4424     Change-Id: Ic4d250a1697047386989f73f058b0abc8b55627b
 4425     Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru>
 4426     Updates: bz#1193929
 4427 
 4428 commit 33453ec69af1070dd5e15dd99c14ebb43c4e88c2
 4429 Author: Yaniv Kaul <ykaul@redhat.com>
 4430 Date:   Fri Dec 27 21:30:21 2019 +0200
 4431 
 4432     posix-inode-fd-ops.c: fix a compiler warning - real_path may be NULL
 4433     
 4434     Fixes the following:
 4435     posix-inode-fd-ops.c: In function ‘posix_common_removexattr’:
 4436     ../../../../libglusterfs/src/glusterfs/logging.h:231:9: warning: ‘%s’ directive argument is null [-Wformat-overflow=]
 4437       231 |         _gf_msg(dom, __FILE__, __FUNCTION__, __LINE__, level, errnum, 0,       \
 4438           |         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 4439       232 |                 msgid, ##fmt);                                                 \
 4440           |                 ~~~~~~~~~~~~~
 4441     posix-inode-fd-ops.c:4401:13: note: in expansion of macro ‘gf_msg’
 4442      4401 |             gf_msg(this->name, GF_LOG_WARNING, errno, P_MSG_FDSTAT_FAILED,
 4443           |             ^~~~~~
 4444     posix-inode-fd-ops.c:4402:47: note: format string is defined here
 4445      4402 |                    "fdstat operaton failed on %s", real_path);
 4446           |                                               ^~
 4447     
 4448     Change-Id: I3f2278d3a05a569dde257d66f44a8655491b4013
 4449     updates: bz#1193929
 4450     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 4451 
 4452 commit ec3df84dcfd7ccda0a18fa75e3b425c090209adf
 4453 Author: Shwetha K Acharya <sacharya@redhat.com>
 4454 Date:   Thu Nov 14 12:37:24 2019 +0530
 4455 
 4456     features/changelog: Optimization in changelog
 4457     
 4458     Problem: Currently changelog is written in one directory, which over
 4459     time, results in very large changelog files.
 4460     
 4461     Solution: Seperate directory under the changelogs directory
 4462     is created on daily basis following the format year/month/day.
 4463     
 4464     Updates: #154
 4465     Change-Id: I1cdabe33728a0ba1f298c8908bd8c323b1871bda
 4466     Signed-off-by: Shwetha K Acharya <sacharya@redhat.com>
 4467 
 4468 commit ca3e5905ac02fb9c373ac3de10b44f061d04cd6f
 4469 Author: Mohit Agrawal <moagrawal@redhat.com>
 4470 Date:   Sun Oct 20 22:01:01 2019 +0530
 4471 
 4472     posix: Avoid diskpace error in case of overwriting the data
 4473     
 4474     Problem: Sometime fops like posix_writev, posix_fallocate, posix_zerofile
 4475              failed and throw error ENOSPC if storage.reserve threshold limit
 4476              has reached even fops is overwriting the data
 4477     
 4478     Solution: Retry the fops in case of overwrite if diskspace check
 4479               is failed
 4480     
 4481     Credits: kinsu <vpolakis@gmail.com>
 4482     Change-Id: I987d73bcf47ed1bb27878df40c39751296e95fe8
 4483     Updates: #745
 4484     Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
 4485 
 4486 commit e21a5a1b1edc5f0ddea54be3dd10869de517b9e6
 4487 Author: Yaniv Kaul <ykaul@redhat.com>
 4488 Date:   Sun Dec 29 11:03:40 2019 +0200
 4489 
 4490     posix-entry-ops.c: remove some tier related code
 4491     
 4492     Remove TIER_LINKFILE_GFID related code from posix
 4493     
 4494     Tier xlator was removed, but there are some code related to it scattered
 4495     around in DHT and Posix xlators. Remove some of it.
 4496     
 4497     Change-Id: I3a878b31ed4a045ed419f936aa1d567ded1a273f
 4498     updates: bz#1193929
 4499     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 4500 
 4501 commit 42f484dcecd9942611396d9bd2ad3a39019b0e1f
 4502 Author: Mohammed Rafi KC <rkavunga@redhat.com>
 4503 Date:   Wed Dec 18 13:16:39 2019 +0530
 4504 
 4505     mgmt/brick-mux: Avoid sending two response when attach is failed.
 4506     
 4507     We were sending two response back to glusterd when an attach is
 4508     failed. One from the handler function glusterfs_handle_attach and
 4509     another from rpcsvc_check_and_reply_error. It was causing problems
 4510     like ref leaks, transport disconnect etc.
 4511     
 4512     Change-Id: I3bb5b59959530760b568d52becb519499b3dcd2b
 4513     updates: bz#1785143
 4514     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
 4515 
 4516 commit 745f6534c514010cc69e2306e1621e9d14ff5dba
 4517 Author: Dmitry Antipov <dmantipov@yandex.ru>
 4518 Date:   Thu Dec 26 13:50:20 2019 +0300
 4519 
 4520     transport/socket: destroy notify mutex and condition variable
 4521     
 4522     Change-Id: Id74f829dc5c6a30d19e3c3ef42bcb938afc0d8e4
 4523     Updates: bz#1430623
 4524     Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru>
 4525 
 4526 commit 9937b9857472c346c071c3850b032ae9237721e5
 4527 Author: Dmitry Antipov <dmantipov@yandex.ru>
 4528 Date:   Thu Dec 26 14:22:56 2019 +0300
 4529 
 4530     socket: fix typos and drop unused members/options
 4531     
 4532     Consistently fix 'configued' -> 'configured' typo, remove useless
 4533     members from 'socket_private_t' and unused 'transport.socket.lowlat'
 4534     option. Adjust tests as well.
 4535     
 4536     Change-Id: I285be196457763aec16b184acd26b90623074dec
 4537     Updates: bz#1193929
 4538     Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru>
 4539 
 4540 commit f2a86ddf9331cd7387c44ba03307adac43575fc4
 4541 Author: Sunny Kumar <sunkumar@redhat.com>
 4542 Date:   Fri Dec 27 13:26:04 2019 +0000
 4543 
 4544     extras: enable log rotation for USS logs
 4545     
 4546     Added logrotate support for user serviceable snapshot's logs.
 4547     
 4548     Change-Id: Ic920eaa8ab5e44daf5937a027c6913d7bb26d517
 4549     Fixes: bz#1786722
 4550     Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
 4551 
 4552 commit 7656aec3b9ef60592c8cf251dfb5cdb6088cd328
 4553 Author: Dmitry Antipov <dmantipov@yandex.ru>
 4554 Date:   Thu Dec 26 15:25:35 2019 +0300
 4555 
 4556     Avoid buffer overwrite due to uuid_utoa() misuse
 4557     
 4558     Code like:
 4559     
 4560     f(..., uuid_utoa(x), uuid_utoa(y));
 4561     
 4562     is not valid (causes undefined behaviour) because uuid_utoa()
 4563     uses the only static thread-local buffer which will be overwritten
 4564     by the subsequent call. All such cases should be converted to use
 4565     uuid_utoa_r() with explicitly specified buffer.
 4566     
 4567     Change-Id: I5e72bab806d96a9dd1707c28ed69ca033b9c8d6c
 4568     Updates: bz#1193929
 4569     Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru>
 4570 
 4571 commit f5aafeffed3a5363ff200e9300dec5855f680d9c
 4572 Author: yatipadia <ypadia@redhat.com>
 4573 Date:   Thu Dec 12 14:40:03 2019 +0530
 4574 
 4575     xlator/bit-rot-stub: structure logging
 4576     
 4577     convert gf_msg() to gf_smsg()
 4578     
 4579     Change-Id: I3072ba4305b720bea053688d140660d7d5c2d267
 4580     Updates: #657
 4581     Signed-off-by: yatipadia <ypadia@redhat.com>
 4582 
 4583 commit 77c48f77a9e7bfa202f9a211bacd7e7a9f88ae6b
 4584 Author: Xavi Hernandez <xhernandez@redhat.com>
 4585 Date:   Thu Dec 19 11:58:54 2019 +0100
 4586 
 4587     timer: fix event destruction race
 4588     
 4589     In current timer implementation, each event has an absolute time at which
 4590     it will be fired. When the first timer of the queue has not elapsed yet,
 4591     a pthread_cond_timedwait() is used to wait until the expected time.
 4592     
 4593     Apparently that's fine. However the time passed to that function was a
 4594     pointer to the timespec structure contained in the event itself. This is
 4595     problematic because of how pthread_cond_timedwait() works internally.
 4596     
 4597     Simplifying a bit, pthread_cond_timedwait() basically queues itself as a
 4598     waiter for the given condition variable and releases the mutex. Then it
 4599     does the timed wait using the passed value.
 4600     
 4601     With that in mind, the follwing case is possible:
 4602     
 4603        Timer Thread                            Other Thread
 4604        ------------                            ------------
 4605     
 4606     					   gf_timer_call_cancel()
 4607        pthread_mutex_lock()                    |
 4608                                                + pthread_mutex_lock()
 4609        event = current_event()                 |
 4610        pthread_cond_timedwait(&event->at)      |
 4611        + pthread_mutex_unlock()                |
 4612        |                                       + remove_event()
 4613        |                                       + destroy_event()
 4614        + timed_wait(&event->at)
 4615     
 4616     As we can see, the time is used after it has been destroyed, which means
 4617     we have a use-after-free problem.
 4618     
 4619     This patch fixes the problem by copying the time to a local variable
 4620     before calling pthread_cond_timedwait()
 4621     
 4622     Change-Id: I0f4e8eded24fe3a1276dc75c6cf093bae973d26b
 4623     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
 4624     Fixes: bz#1785208
 4625 
 4626 commit 6350695dccfafdd6ab22be3509ade549a34db98b
 4627 Author: Sanju Rakonde <srakonde@redhat.com>
 4628 Date:   Tue Dec 24 13:53:16 2019 +0530
 4629 
 4630     Revert "store.c: remove a sys_stat() call - just open the file"
 4631     
 4632     This reverts commit da735ad9b58b76d27c580a9feb63f14b2cc0669d.
 4633     
 4634     This commit is changing the behaviour of glusterd during its
 4635     initialization. The removed piece of code checks whether the
 4636     file mentioned in the path exists or not, if not it returns
 4637     -1. This return value is considered for the sub sequent
 4638     operations. By removing this code, we always return 0 and
 4639     some part of the code will not be executed in the following
 4640     code path.
 4641     
 4642     glusterd_options_init() -> glusterd_store_retrieve_options() ->
 4643     gf_store_handle_retrieve()
 4644     
 4645     In glusterd_options_init(), we are not executing any code
 4646     if the ret value is 0.
 4647     
 4648         ret = glusterd_store_retrieve_options(this);
 4649         if (ret == 0) {
 4650             goto out;
 4651         }
 4652     
 4653     as we are not executing any code if the return value is 0,
 4654     we are unable to enable the brick-multiplex feature using
 4655     volume set.
 4656     
 4657     fixes: bz#1786459
 4658     Change-Id: I4f52b7b2ea8adb8df5087bb96927e9bb7db5b491
 4659 
 4660 commit 28d87d5d833f2de320169aa766feaccc0bd98262
 4661 Author: Barak Sason <bsasonro@redhat.com>
 4662 Date:   Thu Oct 24 16:28:31 2019 +0300
 4663 
 4664     libglusterfs/logger - code cleaning
 4665     
 4666     Fixes: bz#1765186
 4667     
 4668     1 - Removed code duplication of 'log_levles' array
 4669     2 - Declared all static methods at the begining of the .c file
 4670     3 - Removed unused structs that existed in the .c file
 4671     
 4672     Change-Id: I37330cb133fc65deac69b7ca60779f70501c24ea
 4673     Signed-off-by: Barak Sason <bsasonro@redhat.com>
 4674 
 4675 commit e879ff1d4a3a0d795da2575406414eaeb28975f4
 4676 Author: Yaniv Kaul <ykaul@redhat.com>
 4677 Date:   Tue Dec 3 15:28:13 2019 +0200
 4678 
 4679     md-cache.c: move cache-swift-metadata to off by default
 4680     
 4681     This causes mdc_xattr_list_populate() NOT to add "user.swift.metadata"
 4682     as an xattr in the list of attrs we look at in some paths of the code.
 4683     
 4684     This is documented @
 4685     https://github.com/gluster/glusterfs/issues/775
 4686     
 4687     Change-Id: Ie3d676c74a2f333beeacc302e253efe9f9942d1a
 4688     updates: bz#1193929
 4689     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 4690 
 4691 commit da735ad9b58b76d27c580a9feb63f14b2cc0669d
 4692 Author: Yaniv Kaul <ykaul@redhat.com>
 4693 Date:   Mon Nov 25 15:45:27 2019 +0200
 4694 
 4695     store.c: remove a sys_stat() call - just open the file
 4696     
 4697     I don't see the value on performing a sys_stat() on a path
 4698     instead of just trying to open the file.
 4699     
 4700     Change-Id: Ib3f88d8c45aea091eafea17c4be02e2acdb713f1
 4701     updates: bz#1193929
 4702     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 4703 
 4704 commit 82ba1cb8ec8cdbc02ad69a2ab4c6cf569503ab34
 4705 Author: yatipadia <ypadia@redhat.com>
 4706 Date:   Mon Nov 4 21:04:45 2019 +0530
 4707 
 4708     afr_inode/xlator: structure logging
 4709     
 4710     convert gf_msg() into gf_smsg()
 4711     
 4712     Change-Id: I8f5b7bbb9caa78902b06f67257502b67adab7405
 4713     Updates: #657
 4714     Signed-off-by: yatipadia <ypadia@redhat.com>
 4715 
 4716 commit 3389bc58ea9773a66ed4bc7bd96b93dc1c97e4cc
 4717 Author: Raghavendra G <rgowdapp@redhat.com>
 4718 Date:   Wed Nov 20 22:31:48 2019 +0530
 4719 
 4720     tests/fuse/bug-965974.t: turn off md-cache, to fix failure of the test
 4721     
 4722     Imagine the following set of operations:
 4723     1. touch $M0/file
 4724     2. ln $M0/file $M0/file.lnk
 4725     3. rm $M1/file
 4726     4. wait for md-cache-timeout
 4727     5. stat $M0/file.lnk $M0/file
 4728     
 4729     stat on $M0/file in step 5 succeeds when md-cache-timeout is non-zero
 4730     even though it was removed from $M1. The reason is
 4731     1. fuse-bridge on $M0 would resolve both file and file.lnk to same
 4732        inode. This inode i1 is cached in md-cache.
 4733     2. After md-cache-timeout lookup on $M0/file.lnk would be sent to
 4734        backend. This lookup will be successful as file.lnk is present on
 4735        backend and would refresh the md-cache.
 4736     3. The lookup on $M0/file sent within md-cache-timeout after lookup on
 4737        $M0/file.lnk would be sent with i1. Since i1 was refreshed by
 4738        lookup on $M0/file.lnk, the cache is deemed valid and md-cache
 4739        responds lookup on $M0/file as success
 4740     
 4741     To fix this failure we can either:
 4742     1. remove lookup on $M0/file.lnk, so that lookup on $M0/file is
 4743        actually sent to backend.
 4744     2. turn off md-cache
 4745     
 4746     This patch chooses option 2.
 4747     
 4748     credits: Csaba Henk <csaba@redhat.com>
 4749     
 4750     Change-Id: I352c2acd377fe10c4bdf3b6e53c1de86a4e544c7
 4751     Signed-off-by: Raghavendra G <rgowdapp@redhat.com>
 4752     Updates: bz#1756900
 4753 
 4754 commit d8adc66b79de3f21672884539a65327614e0c6bb
 4755 Author: Dmitry Antipov <dmantipov@yandex.ru>
 4756 Date:   Thu Dec 5 16:57:28 2019 +0300
 4757 
 4758     glfsheal: install as architecture-dependent binary helper
 4759     
 4760     Since glfsheal is a binary helper which is not intented to be
 4761     invoked directly, install it under %{libexecdir}/glusterfs rather
 4762     than %{sbindir}, adjust invocation from CLI code and RPM spec.
 4763     
 4764     Credits: Ravishankar N <ravishankar@redhat.com>.
 4765     Fixes: bz#1780190
 4766     
 4767     Change-Id: I4b41892d96b89c24a332470ac8c1e82f6795159a
 4768     Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru>
 4769 
 4770 commit bc6e206c6d29234a0b167b142d20c62ffd529f15
 4771 Author: Yaniv Kaul <ykaul@redhat.com>
 4772 Date:   Mon Oct 28 21:29:50 2019 +0200
 4773 
 4774     cli-rpc-ops.c: cleanups
 4775     
 4776     1. Move functions and structs to static
 4777     2. Use dictionary functions with fixed key length.
 4778     3. Reduce key length when not needed.
 4779     4. Use const char* for some messages.
 4780     5. Use fixed strings for some logs which is repeated in the code.
 4781     6. Remove redundant checks. Specifically, cli_to_glusterd() does
 4782     NULL checks already, so no need to do it before calling it.
 4783     7. Aligned some messages - not sure why they were cut over several
 4784     lines, but it made grep on the code harder.
 4785     8. Move dictionary fetching of values closer to where they are
 4786     actually used.
 4787     
 4788     Overall, object size is ~4 smaller, hopefully without functional changes.
 4789     
 4790     There's more that can be done, but as this is a very long (>10K lines)
 4791     file, I think it's enough for one change.
 4792     Specifically, some functions fetch values from the dictionary
 4793     without using it - this is a bit of a waste.
 4794     Filed https://github.com/gluster/glusterfs/issues/753 about it.
 4795     
 4796     Change-Id: I31f88d94ab25398e00aef2ea84a8c4af9383b75b
 4797     updates: bz#1193929
 4798     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 4799 
 4800 commit 8c6a19783e149dda1ab9143f48badfa617478e6e
 4801 Author: Barak Sason Rofman <bsasonro@redhat.com>
 4802 Date:   Tue Nov 5 15:54:29 2019 +0200
 4803 
 4804     glusterd: refactoring long method
 4805     
 4806     - Refactored set_fuse_mount_options(...) in order to shorten it.
 4807     - Removed dead code and moved some method to it's apropriate
 4808     location.
 4809     - Converted loggin in set_fuse_mount_options(...) to structured logs
 4810     
 4811     fixes: bz#1768896
 4812     Change-Id: If865833d4c60d517da202871978691ef21235fe4
 4813     Signed-off-by: Barak Sason Rofman <bsasonro@redhat.com>
 4814 
 4815 commit a7eeab9b497724cdc4de5a82b880ba824a9d8358
 4816 Author: Sunny Kumar <sunkumar@redhat.com>
 4817 Date:   Mon Dec 16 12:52:41 2019 +0000
 4818 
 4819      test: fix non-root test case for geo-rep
 4820     
 4821     Problem:
 4822     On a freshly installed system non-root geo-rep test case gets blocked.
 4823     
 4824     Solution:
 4825     
 4826     On a freshly installed system, the remote key need to be accepted automatically by ssh-copy-id.
 4827     
 4828     Credits: M. Scherer <mscherer@redhat.com>
 4829     
 4830     Change-Id: I5077f99a6681660f7e3e84c25ef216f521b7c29c
 4831     Fixes: bz#1779742
 4832     Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
 4833 
 4834 commit 194e4216c8e30612f96f8f3712fdb3f1c96b3a08
 4835 Author: Yaniv Kaul <ykaul@redhat.com>
 4836 Date:   Tue Dec 3 12:57:13 2019 +0200
 4837 
 4838     [RFC]#ifdef gNFS related code if we are not compiling gNFS
 4839     
 4840     If we are not compiling gNFS (--enable-gnfs is not given in the
 4841     ./configure script params), there is little point in compiling code
 4842     that is related to it.
 4843     
 4844     This patch tries to eliminate it.
 4845     My hope (and it's not clear from the code ) is that I did not break
 4846     the NFS Ganesha support as well.
 4847     
 4848     Other than that, tried to compile with and without anad it looks sane.
 4849     
 4850     Change-Id: I8d6c98066b9fceab4ec10fc6f5e81ab069e853bd
 4851     updates: bz#1193929
 4852     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 4853 
 4854 commit 6f6b253deed134c5fee37e1e3d2fd44b18e1db6e
 4855 Author: Sanju Rakonde <srakonde@redhat.com>
 4856 Date:   Tue Dec 17 15:52:30 2019 +0530
 4857 
 4858     glusterd: unlink the file after killing the process
 4859     
 4860     In glusterd_proc_stop(), after killing the pid
 4861     we should remove the pidfile.
 4862     
 4863     fixes: bz#1784375
 4864     
 4865     Change-Id: Ib6367aed590932c884b0f6f892fc40542aa19686
 4866     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
 4867 
 4868 commit 8537d5bda5fa5d52840e24884eaa3f2bcfcd6bd9
 4869 Author: Yaniv Kaul <ykaul@redhat.com>
 4870 Date:   Thu Dec 12 08:12:44 2019 +0200
 4871 
 4872     call-stub.c: remove redundant validation
 4873     
 4874     GF_VALIDATE_OR_GOTO("call-stub", frame, out) is removed.
 4875     It's not needed, since we are testing via the creation of a stub,
 4876     in stub_new() that frame is not NULL.
 4877     
 4878     Removed all redundant checks, which saves  >10% in object size.
 4879     
 4880     Change-Id: If4ce3a3dc506a5e65e98b59fa51b1fdd389a85ba
 4881     updates: bz#1193929
 4882     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 4883 
 4884 commit f6e2958543246d547f1c3c3b3b080d435b4c9296
 4885 Author: kshithijiyer <kshithij.ki@gmail.com>
 4886 Date:   Sat Nov 30 15:25:11 2019 +0530
 4887 
 4888     [Cli] Removing old log rotate command.
 4889     
 4890     The old command for log rotate is still present removing
 4891     it completely. Also adding testcase to test the
 4892     log rotate command with both the old as well as the new command
 4893     and fixing testcase which use the old syntax to use the new
 4894     one.
 4895     
 4896     Code to be removed:
 4897     1. In cli-cmd-volume.c from struct cli_cmd volume_cmds[]:
 4898     {"volume log rotate <VOLNAME> [BRICK]", cli_cmd_log_rotate_cbk,
 4899      "rotate the log file for corresponding volume/brick"
 4900      " NOTE: This is an old syntax, will be deprecated from next release."},
 4901     
 4902     2. In cli-cmd-volume.c from cli_cmd_log_rotate_cbk():
 4903      ||(strcmp("rotate", words[2]) == 0)))
 4904     
 4905     3. In cli-cmd-parser.c from cli_cmd_log_rotate_parse()
 4906     if (strcmp("rotate", words[2]) == 0)
 4907        volname = (char *)words[3];
 4908     else
 4909     
 4910     fixes: bz#1750387
 4911     Change-Id: I56e4d295044e8d5fd1fc0d848bc87e135e9e32b4
 4912     Signed-off-by: kshithijiyer <kshithij.ki@gmail.com>
 4913 
 4914 commit ff7865be148d5f4f22861fe93710c0b97d24f81f
 4915 Author: Barak Sason Rofman <bsasonro@redhat.com>
 4916 Date:   Tue Nov 26 12:36:16 2019 +0200
 4917 
 4918     DHT - Reduce methods scope (dht-common.c)
 4919     
 4920     Methods that should have been static were defined as global, and the
 4921     other way around.
 4922     This patch fixes the issue in order to enforce encapsulation.
 4923     
 4924     updates: bz#1776757
 4925     
 4926     Change-Id: I3eb5781849c5e597c1dd347e03f356c00db62a39
 4927     Signed-off-by: Barak Sason Rofman <bsasonro@redhat.com>
 4928 
 4929 commit 49f524def676e62e0f0cfb8a674bc28923c28f13
 4930 Author: Yaniv Kaul <ykaul@redhat.com>
 4931 Date:   Tue Oct 22 15:41:41 2019 +0300
 4932 
 4933     dht-common.h/dht-helper.c: exctract the LOCK() from DHT_UPDATE_TIME
 4934     
 4935     Currently, the code (and only place) that is using this macro is
 4936     in dht_inode_ctx_time_update() where it is called 3 times in a row,
 4937     which is essentially 3 cycles of LOCK/UNLOCK on the same lock.
 4938     
 4939     Instead, extract the LOCK()/UNLOCK() part of the macro
 4940     and wrap those calls with it.
 4941     
 4942     Change-Id: I6312b985e3d97517857b55f342440accc4063db6
 4943     updates: bz#1193929
 4944     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 4945 
 4946 commit 5ab835004371f2165d1c3f335aa0fd777c9f8c20
 4947 Author: Andrew Miloradovsky <andrew@interpretmath.pw>
 4948 Date:   Mon Dec 2 17:36:48 2019 +0300
 4949 
 4950     rpc: define xdr_sizeof regardless of whether IPv6 is the default
 4951     
 4952     Change-Id: I4f20f376d82b28e1c572c0fd0b6cd38e97b133da
 4953     Fixes: bz#1780260
 4954 
 4955 commit b202e0e01d0eb8085eb8006dfa77b7b4b06978e6
 4956 Author: Xavi Hernandez <xhernandez@redhat.com>
 4957 Date:   Wed Dec 11 18:21:14 2019 +0100
 4958 
 4959     socket: fix error handling
 4960     
 4961     When __socket_proto_state_machine() detected a problem in the size of
 4962     the request or it couldn't allocate an iobuf of the requested size, it
 4963     returned -ENOMEM (-12). However the caller was expecting only -1 in
 4964     case of error. For this reason the error passes undetected initially,
 4965     adding back the socket to the epoll object. On further processing,
 4966     however, the error is finally detected and the connection terminated.
 4967     Meanwhile, another thread could receive a poll_in event from the same
 4968     connection, which could cause races with the connection destruction.
 4969     When this happened, the process crashed.
 4970     
 4971     To fix this, all error detection conditions have been hardened to be
 4972     more strict on what is valid and what not. Also, we don't return
 4973     -ENOMEM anymore. We always return -1 in case of error.
 4974     
 4975     An additional change has been done to prevent destruction of the
 4976     transport object while it may still be needed.
 4977     
 4978     Change-Id: I6e59cd81cbf670f7adfdde942625d4e6c3fbc82d
 4979     Fixes: bz#1782495
 4980     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
 4981 
 4982 commit d7e049160a9dea988ded5816491c2234d40ab6b3
 4983 Author: Ravishankar N <ravishankar@redhat.com>
 4984 Date:   Thu Nov 7 15:18:30 2019 +0530
 4985 
 4986     afr: make heal info lockless
 4987     
 4988     Changes in locks xlator:
 4989     Added support for per-domain inodelk count requests.
 4990     Caller needs to set GLUSTERFS_MULTIPLE_DOM_LK_CNT_REQUESTS key in the
 4991     dict and then set each key with name
 4992     'GLUSTERFS_INODELK_DOM_PREFIX:<domain name>'.
 4993     In the response dict, the xlator will send the per domain count as
 4994     values for each of these keys.
 4995     
 4996     Changes in AFR:
 4997     Replaced afr_selfheal_locked_inspect() with afr_lockless_inspect(). Logic has
 4998     been added to make the latter behave same as the former, thus not
 4999     breaking the current heal info output behaviour.
 5000     
 5001     fixes: bz#1774011
 5002     Change-Id: Ie9e83c162aa77f44a39c2ba7115de558120ada4d
 5003     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
 5004 
 5005 commit 9e79604518dcf75a9f88a2444c177e97fd13067c
 5006 Author: Xie Changlong <xiechanglong@cmss.chinamobile.com>
 5007 Date:   Tue Dec 3 13:53:34 2019 +0800
 5008 
 5009     contributors: map domain names for CMSS
 5010     
 5011     updates: bz#1193929
 5012     Change-Id: Iaf1e1e3f787855ede1e94101ec0364084e534d61
 5013     Signed-off-by: Xie Changlong <xiechanglong@cmss.chinamobile.com>
 5014 
 5015 commit 86f54c6f82eae33f4070705678fc3b0ff6b7e3d5
 5016 Author: Xie Changlong <xiechanglong@cmss.chinamobile.com>
 5017 Date:   Mon Dec 2 13:20:02 2019 +0800
 5018 
 5019     fuse: add missing unlock
 5020     
 5021     updates: bz#1193929
 5022     Change-Id: I50f75f730ea6970e99347fcee661ce9dc8477725
 5023     Signed-off-by: Xie Changlong <xiechanglong@cmss.chinamobile.com>
 5024 
 5025 commit 8030f9c0f092170ceb50cedf59b9c330022825b7
 5026 Author: Mohit Agrawal <moagrawal@redhat.com>
 5027 Date:   Tue Dec 10 08:35:23 2019 +0530
 5028 
 5029     rpc: event_slot_alloc converted infinite loop after reach slot_used to 1024
 5030     
 5031     Problem: In the commit faf5ac13c4ee00a05e9451bf8da3be2a9043bbf2 missed one
 5032              condition to come out from the loop so after reach the slot_used to
 5033              1024 loop has become infinite loop
 5034     
 5035     Solution: Correct the code path to avoid the infinite loop
 5036     
 5037     Change-Id: Ia02a109571f0d8cc9902c32db3e9b9282ee5c1db
 5038     Fixes: bz#1781440
 5039     Credits: Xavi Hernandez <xhernandez@redhat.com>
 5040     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 5041 
 5042 commit f8f4efef77c177900923f5b29ddb707fb3f07720
 5043 Author: Barak Sason Rofman <bsasonro@redhat.com>
 5044 Date:   Sun Dec 1 12:26:53 2019 +0200
 5045 
 5046     afr/self-heald - Missing error logs
 5047     
 5048     As a follow up on https://review.gluster.org/#/c/glusterfs/+/23749/,
 5049     adding error logging for the entire method.
 5050     In addition, converted logging to structured logging in the method.
 5051     
 5052     Fixes: bz#1778457
 5053     Change-Id: I1f412159e6849d6f6ddbde53ec4a85ad709bbdf4
 5054     Signed-off-by: Barak Sason Rofman <bsasonro@redhat.com>
 5055 
 5056 commit ba407a67fc0895c970c327361110a960af254401
 5057 Author: Anoop C S <anoopcs@redhat.com>
 5058 Date:   Fri Nov 22 17:36:55 2019 +0530
 5059 
 5060     extras/hooks: Remove smb.conf parameter allowing guest access
 5061     
 5062     Change-Id: I88f494f16153d27ab6e2f2faf4d557e075671b10
 5063     Fixes: bz#1775612
 5064     Signed-off-by: Anoop C S <anoopcs@redhat.com>
 5065 
 5066 commit 990d9fb3973cc3d3e211f0b4eebd8e909efd8028
 5067 Author: Barak Sason Rofman <bsasonro@redhat.com>
 5068 Date:   Tue Dec 3 12:13:16 2019 +0200
 5069 
 5070     cloudsync - fixing a coverity issue
 5071     
 5072     A negative value was being passed where it shouldn't have been passed
 5073     to. Modified code so only a positive value will be passed.
 5074     
 5075     fixes: CID#1274209
 5076     Change-Id: I5452ddded77664fdd8b5c975af6bc77806a7ffb1
 5077     updates: bz#789278
 5078 
 5079 commit 400cf809ed24ffb506fed6edd510643be406e3ea
 5080 Author: Kotresh HR <khiremat@redhat.com>
 5081 Date:   Mon Nov 18 05:24:33 2019 -0500
 5082 
 5083     ctime: Fix ctime inconsisteny with utimensat
 5084     
 5085     Problem:
 5086     When touch is used to create a file, the ctime is not matching
 5087     atime and mtime which ideally should match. There is a difference
 5088     in nano seconds.
 5089     
 5090     Cause:
 5091     When touch is used modify atime or mtime to current time (UTIME_NOW),
 5092     the current time is taken from kernel. The ctime gets updated to current
 5093     time when atime or mtime is updated. But the current time to update
 5094     ctime is taken from utime xlator. Hence the difference in nano seconds.
 5095     
 5096     Fix:
 5097     When utimesat uses UTIME_NOW, use the current time from kernel.
 5098     
 5099     fixes: bz#1773530
 5100     Change-Id: I9ccfa47dcd39df23396852b4216f1773c49250ce
 5101     Signed-off-by: Kotresh HR <khiremat@redhat.com>
 5102 
 5103 commit d30c0b5fa4dba283587cb77a95a79b395cad8e1b
 5104 Author: Andrew Miloradovsky <andrew@interpretmath.pw>
 5105 Date:   Mon Dec 2 13:23:16 2019 +0300
 5106 
 5107     git-ignore .dirstamp
 5108     
 5109     Change-Id: I8a0191252c017bda9e9d42f9622504698bf6e7da
 5110     Fixes: bz#1776892
 5111 
 5112 commit 75205f901c06f39fdd41ebd13303e0a70a460041
 5113 Author: N Balachandran <nbalacha@redhat.com>
 5114 Date:   Tue Dec 3 10:29:37 2019 +0530
 5115 
 5116     cluster/dht: Add comments to code
 5117     
 5118     Change-Id: Ieb7531af19ae89fb8a8387e81663c7f157b10c02
 5119     Updates: bz#1765421
 5120     Signed-off-by: N Balachandran <nbalacha@redhat.com>
 5121 
 5122 commit 99044a5cedcff9a9eec40a07ecb32bd66271cd02
 5123 Author: HuangShujun <549702281@qq.com>
 5124 Date:   Thu Dec 5 10:07:10 2019 +0200
 5125 
 5126     To fix readdir-ahead memory leak
 5127     
 5128     Glusterfs client process has memory leak if create serveral files under one folder, and delete the folder.
 5129     According to statedump, the ref counts of readdir-ahead is bigger than zero in the inode table. Readdir-ahead get parent inode by inode_parent in rda_mark_inode_dirty when each rda_writev_cbk,the inode ref count of parent folder will be increased in inode_parent, but readdir-ahead do not unref it later.
 5130     The correction is unref the parent inode at the end of rda_mark_inode_dirty
 5131     
 5132     Fixes: bz#1779055
 5133     Signed-off-by: HuangShujun <549702281@qq.com>
 5134     
 5135     Change-Id: Iee68ab1089cbc2fbc4185b93720fb1f66ee89524
 5136 
 5137 commit b20b60027c7212d7ecba96a7d75b5f5cbcc08989
 5138 Author: Ashish Pandey <aspandey@redhat.com>
 5139 Date:   Fri Sep 6 14:47:04 2019 +0530
 5140 
 5141     cluster/ec: Scripts to find and reset files which can be healed.
 5142     
 5143     gfid_needing_heal_parallel.sh - Finds out if a file is healable or not.
 5144     
 5145     correct_pending_heals.sh - Makes necessary changes in xattrs to enable
 5146     heal for those files which could be healed.
 5147     
 5148     fixes: #723
 5149     Change-Id: I38177888df0fda9486343ee546dc02836b06a5fc
 5150 
 5151 commit ff53751b37656d6a530c5a347bb1aa84649fff6d
 5152 Author: Sanju Rakonde <srakonde@redhat.com>
 5153 Date:   Tue Oct 22 19:03:13 2019 +0530
 5154 
 5155     core: fix spelling mistakes
 5156     
 5157     fixes: bz#1728554
 5158     
 5159     credits: Patrick Matthäi <pmatthaei@debian.org>
 5160     
 5161     Change-Id: Id08dabf54a529dbb86666b544b1f0859aab75aac
 5162     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
 5163 
 5164 commit 3199f8759ebfcc31ce158201c3c25d9a738479f8
 5165 Author: Dmitry Antipov <dmantipov@yandex.ru>
 5166 Date:   Wed Dec 4 15:46:16 2019 +0300
 5167 
 5168     core: avoid NULL pointer dereference
 5169     
 5170     Since low-level __gf_xxx allocation function may be called before
 5171     ctx is initialized, add extra check for NULL where appropriate.
 5172     
 5173     Fixes: bz#1776784
 5174     Change-Id: I3127fa4b93f8e3e5846106aadcfed1baa27ac43f
 5175     Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru>
 5176 
 5177 commit dc757b2c97f5917885db76c62fa4fb429ae7b9ab
 5178 Author: Sanju Rakonde <srakonde@redhat.com>
 5179 Date:   Tue Nov 26 17:03:44 2019 +0530
 5180 
 5181     glusterd: set xaatrs after checking the brick order
 5182     
 5183     Problem: When volume creation fails complaining about
 5184     the bricks from same hosts for replica volumes, the
 5185     bricks can't be re-used to create any volume without
 5186     using force at the end. It says, brick is already part
 5187     of a volume.
 5188     
 5189     Reason: When volume create opeartion issued, we set
 5190     xattrs on the bricks. If the transaction fails in later
 5191     checks, the xattrs will remain on the brick. When the
 5192     brick is re-used, by looking at the xattrs, glusterd
 5193     thinks it is already part of volume.
 5194     
 5195     Solution: Check the brick order for replica and disperse
 5196     volumes before setting the xattrs.
 5197     
 5198     fixes: bz#1776801
 5199     
 5200     Change-Id: I44a971b37f520e5a20dc9fad6520286d315063b9
 5201     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
 5202 
 5203 commit 6015ab7d8b37388e836f36a315af54f019839d91
 5204 Author: Sanju Rakonde <srakonde@redhat.com>
 5205 Date:   Mon Nov 25 16:35:42 2019 +0530
 5206 
 5207     glusterd: start glusterd automatically on abnormal shutdown
 5208     
 5209     If glusterd crashes or goes down abnormally, systemd should
 5210     automatically bring the glusterd up.
 5211     
 5212     With this change, systemd brings glusterd up for atmost 3 times
 5213     within time period of 1 hour. If the limit exceeds, we have to
 5214     start the glusterd manually and reset the failure count using
 5215     systemctl reset-failed.
 5216     
 5217     credits: John Strunk <jstrunk@redhat.com>
 5218     
 5219     fixes: bz#1776264
 5220     
 5221     Change-Id: I312d243652fb13ba028814a2ea615b67e3b10b6a
 5222     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
 5223 
 5224 commit d29b8e62a186770d26bf7daf3437d899ae00a0da
 5225 Author: Xie Changlong <xiechanglong@cmss.chinamobile.com>
 5226 Date:   Tue Dec 3 13:39:47 2019 +0800
 5227 
 5228     glusterd-store: fix Coverity CID 1407767
 5229     
 5230     To avoid memory leak.
 5231     
 5232     updates: bz#789278
 5233     Change-Id: I321c65a66c9ac372d059aa3ca89a989c6089979e
 5234     Signed-off-by: Xie Changlong <xiechanglong@cmss.chinamobile.com>
 5235 
 5236 commit c65c0a72f4f1597ffa8bc0f8a8dc83abedf3e099
 5237 Author: Jiffin Tony Thottan <jthottan@redhat.com>
 5238 Date:   Fri Nov 29 17:07:14 2019 +0530
 5239 
 5240     MAINTAINERS: add Xie Changlong and Amar Tumballi as peers for gNFS
 5241     
 5242     * Change status of gNFS from Deprecated to Odd Fixes
 5243     
 5244     More discussion on this @ thread in gluster-devel:
 5245     https://lists.gluster.org/pipermail/gluster-devel/2019-November/056685.html
 5246     
 5247     Change-Id: I119f588f3ad07fa55764ee7cf7db42905e19ee1b
 5248     Updates: bz#1193929
 5249     Signed-off-by: Jiffin Tony Thottan <jthottan@redhat.com>
 5250 
 5251 commit 8483753514393239ba46e11c3a83e7ebff4cc3f4
 5252 Author: Vishal Pandey <vpandey@redhat.com>
 5253 Date:   Thu Nov 21 12:56:34 2019 +0530
 5254 
 5255     man/gluster: Add volume top command to gluster man page
 5256     
 5257     Change-Id: Ib74607d2b2e5a1a0316221f1176a7dcccea632d4
 5258     Fixes: bz#1774866
 5259     Signed-off-by: Vishal Pandey <vpandey@redhat.com>
 5260 
 5261 commit b35f8addc6b582a2498dfae546fbb942964c66b3
 5262 Author: Yaniv Kaul <ykaul@redhat.com>
 5263 Date:   Sun Nov 3 15:21:05 2019 +0200
 5264 
 5265     glusterd-op-sm.c (and others) - improve glusterd_op_stage_set_volume()
 5266     
 5267     Multiple changes to the function in the hope to make it somewhat faster.
 5268     1. Checking for key length against constant strings before calling
 5269     strcmp() to save some calls.
 5270     2. Verifying if a match was already made against the key to reduce
 5271     yet more checks.
 5272     3. Alignment of error message when they can fit on less lines - just
 5273     makes 'grep' on the code for error messages easier and it's more
 5274     readable.
 5275     4. Multiple functions where call _gd_get_vmep() one by one.
 5276     Instead, extracted it to be callable (it was static) and re-used
 5277     its result, instead of calling it again and again.
 5278     5. Removed some unneeded include statement.
 5279     6. Removed redundant null checks.
 5280     
 5281     Hopefully, no functional changes.
 5282     
 5283     Change-Id: Id281224e49adeca6757f96653b4cb13c7c9ba8c9
 5284     updates: bz#1193929
 5285     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 5286 
 5287 commit 1bfb0a750c6fcd0bbf4d68a1890704f0aefe6337
 5288 Author: Amar Tumballi <amar@kadalu.io>
 5289 Date:   Sun Nov 17 11:00:50 2019 +0530
 5290 
 5291     debug/io-stats: add an option to set volume-id
 5292     
 5293     'volume-id' is good to have for a graph for uniquely identifying it.
 5294     Add it to graph->volume_id while generating volfile itself.
 5295     
 5296     This can be further used in many other places.
 5297     
 5298     Updates: #763
 5299     Change-Id: I80516d62d28a284e8ff4707841570ced97a37e73
 5300     Signed-off-by: Amar Tumballi <amar@kadalu.io>
 5301 
 5302 commit 7c1e383876ad15ec6fc93d412311662f42390a11
 5303 Author: Raghavendra G <rgowdapp@redhat.com>
 5304 Date:   Thu Nov 28 10:31:11 2019 +0530
 5305 
 5306     MAINTAINERS: Add Nithya to past contributors section
 5307     
 5308     Change-Id: I18195a61314946930db669fcf72703a7a18f43a1
 5309     Signed-off-by: Raghavendra G <rgowdapp@redhat.com>
 5310     updates: bz#1193929
 5311 
 5312 commit 1e5cc2754d88089be48e68cf70e0e59d6b4c99ca
 5313 Author: Barak Sason Rofman <bsasonro@redhat.com>
 5314 Date:   Sun Nov 24 18:16:54 2019 +0200
 5315 
 5316     cluster/afr - coverity issue fix
 5317     
 5318     Added a log for a failure in order to avoid "unused variable" coverity
 5319     issue.
 5320     
 5321     fixes: CID#1274209
 5322     
 5323     Change-Id: Ibc6b0ab4bdff482096e42e88fd4c8c7eadfeeadb
 5324     Updates: bz#789278
 5325     Signed-off-by: Barak Sason Rofman <bsasonro@redhat.com>
 5326 
 5327 commit 3a8969b852a70bc05887c764474039279ceb23ff
 5328 Author: Ravishankar N <ravishankar@redhat.com>
 5329 Date:   Thu Nov 28 07:19:13 2019 +0000
 5330 
 5331     Revert "afr: make heal info lockless"
 5332     
 5333     This reverts commit fce5f68bc72d448490a0d41be494ac54a9181b3c.
 5334     I merged the wrong patch by mistake! Hence reverting it.
 5335     
 5336     updates: bz#1774011
 5337     Change-Id: Id7d6ed1d727efc02467c8a9aea3374331261ebd5
 5338 
 5339 commit fce5f68bc72d448490a0d41be494ac54a9181b3c
 5340 Author: Ravishankar N <ravishankar@redhat.com>
 5341 Date:   Thu Nov 7 15:18:30 2019 +0530
 5342 
 5343     afr: make heal info lockless
 5344     
 5345     Changes in locks xlator:
 5346     Added support for per-domain inodelk count requests.
 5347     Caller needs to set GLUSTERFS_MULTIPLE_DOM_LK_CNT_REQUESTS key in the
 5348     dict and then set each key with name
 5349     'GLUSTERFS_INODELK_DOM_PREFIX:<domain name>'.
 5350     In the response dict, the xlator will send the per domain count as
 5351     values for each of these keys.
 5352     
 5353     Changes in AFR:
 5354     Replaced afr_selfheal_locked_inspect() with afr_lockless_inspect(). Logic has
 5355     been added to make the latter behave same as the former, thus not
 5356     breaking the current heal info output behaviour.
 5357     
 5358     fixes: bz#1774011
 5359     Change-Id: I9ae08ce768b39aeb6ee230207b5b7fa744176952
 5360     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
 5361 
 5362 commit 8424df172339d3b2b65bc022e5c8f150f8826dba
 5363 Author: Yaniv Kaul <ykaul@redhat.com>
 5364 Date:   Tue Nov 5 20:37:38 2019 +0200
 5365 
 5366     common-utils.c: add gf_strn2boolean() function
 5367     
 5368     The function takes a string and its length and based on it
 5369     returns if it's a boolean. It's identical in functionality
 5370     to gf_string2boolean only with far less string comparisons since
 5371     it takes into account the length of the string.
 5372     
 5373     dict_get_str_boolean() has been converted to use it.
 5374     
 5375     Other cases of gf_string2boolean() across the code base can be
 5376     converted as well, but more importantly, they should be converted
 5377     from dict_get_str() and then calling to gf_string2boolean to
 5378     simply call dict_get_str_boolean(), which would take care of this
 5379     for them.
 5380     
 5381     This is therefore a first step in the conversion.
 5382     
 5383     Change-Id: I9ee93abfc676f6e123a3919d8df8c25e8848b799
 5384     updates: bz#1193929
 5385     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 5386 
 5387 commit 023a565875891b8078d143a25fee9d44de365c88
 5388 Author: root <root@dhcp35-62.lab.eng.blr.redhat.com>
 5389 Date:   Tue Nov 26 10:30:17 2019 +0530
 5390 
 5391     Maintainers: Remove Nithya as maintainer/peer
 5392     
 5393     Change-Id: I9512a70622b51416174ff0936f8d02e3c0178e19
 5394     Updates: bz#1193929
 5395     Signed-off-by: Nithya Balachandran <nbalacha@redhat.com>
 5396 
 5397 commit 87ef6a23dee165b00418a36ad5473f4427928e49
 5398 Author: Yaniv Kaul <ykaul@redhat.com>
 5399 Date:   Mon Nov 25 17:39:07 2019 +0200
 5400 
 5401     glusterfsd.c: remove sys_lstat() call
 5402     
 5403     get_volfp() in glfs.c doesn't use it, so get_volfp() in glusterfsd.c
 5404     can just open the file without the stat call as well, IMHO.
 5405     
 5406     Change-Id: I3cb5bf12a09b5be42aa2ee4f432f8d351eee5b9e
 5407     updates: bz#1193929
 5408     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 5409 
 5410 commit d97f78467849eef92559217b48d2158521cbba5f
 5411 Author: Yaniv Kaul <ykaul@redhat.com>
 5412 Date:   Mon Nov 25 15:37:46 2019 +0200
 5413 
 5414     store.c/glusterd-store.c: remove sys_stat calls
 5415     
 5416     Instead of querying for the file size and allocating a char array
 5417     according to its size, let's just use a fixed size.
 5418     Those calls are not really needed, and are either expensive or
 5419     cached anyway. Since we do dynamic allocation/free, let's just use
 5420     a fixed array instead.
 5421     
 5422     I'll see if there are other sys_stat() calls that are not really
 5423     useful and try to eliminate them in separate patches.
 5424     
 5425     Change-Id: I76b40e78a52ab38f613fc0cdef4be60e6253bf20
 5426     updates: bz#1193929
 5427     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 5428 
 5429 commit d60935d1011e387115e0445629976196f566b3b1
 5430 Author: Amar Tumballi <amar@kadalu.io>
 5431 Date:   Tue Nov 26 12:53:11 2019 +0530
 5432 
 5433     pcsvc: fix subnet_mask_v4 check
 5434     
 5435     The check we had for subnet mask validation wasn't checking in
 5436     proper sequence. Corrected the order of calling `inet_pton()` as
 5437     the fix.
 5438     
 5439     Fixes: #765
 5440     Change-Id: I5d31468eb917aa94cbb85f573b37c60023e9daf3
 5441     Signed-off-by: Amar Tumballi <amar@kadalu.io>
 5442 
 5443 commit 337e6d6b7c6612a4ce360c9e4d1a484436cdef57
 5444 Author: Yaniv Kaul <ykaul@redhat.com>
 5445 Date:   Mon Nov 25 21:00:59 2019 +0200
 5446 
 5447     posix-inode-fd-ops.c: excute sys_fstat() only when needed
 5448     
 5449     It appears that in posix_do_futimes() we may not need to
 5450     unconditionally execute sys_fstat(). Avoid it and use the
 5451     existing stbuf atime and mtime if possible.
 5452     If not, we execute it.
 5453     
 5454     Change-Id: I0bdd471e5c821fcd28f057c75046c673a212d347
 5455     updates: bz#1193929
 5456     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 5457 
 5458 commit b7411e36514f2e564b3808679732ad8a5c1d9efd
 5459 Author: Yaniv Kaul <ykaul@redhat.com>
 5460 Date:   Fri Oct 4 23:14:08 2019 +0300
 5461 
 5462     cli-*: make some structs static to reduce space consumed.
 5463     
 5464     There's a small difference when structs are defined static.
 5465     Whenever possible, define them as such.
 5466     
 5467     Specifically, before:
 5468        text    data     bss     dec     hex filename
 5469         678     216       0     894     37e ./cli/src/cli-cmd-misc.o
 5470      150024    1264      16  151304   24f08 ./cli/src/cli-rpc-ops.o
 5471       71980      64       0   72044   1196c ./cli/src/cli-cmd-parser.o
 5472       66189       4      16   66209   102a1 ./cli/src/cli-xml-output.o
 5473     
 5474     After:
 5475        text	   data	    bss	    dec	    hex	filename
 5476         670	    216	      0	    886	    376	./cli/src/cli-cmd-misc.o
 5477      149848	   1392	     16	 151256	  24ed8	./cli/src/cli-rpc-ops.o
 5478       70346	   1320	      0	  71666	  117f2	./cli/src/cli-cmd-parser.o
 5479       66157	      4	     16	  66177	  10281	./cli/src/cli-xml-output.o
 5480     
 5481     Change-Id: I206bd895290595d79fac7b26eee66f4279b50f92
 5482     updates: bz#1193929
 5483     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 5484 
 5485 commit a8d54730bba5050ad53cfa46477e698d565e05b1
 5486 Author: Yaniv Kaul <ykaul@redhat.com>
 5487 Date:   Mon Nov 18 22:20:01 2019 +0200
 5488 
 5489     glusterfsd-mgmt.c: move INFO log outside a LOCK
 5490     
 5491     In glusterfs_handle_attach() we can:
 5492     1. Move an INFO level to be executed before the LOCK.
 5493     2. Skip the LOCK altogether, if there's no active graph.
 5494     
 5495     I hope it's safe - I've seen that in other functions
 5496     you could look at ctx->active outside of a lock.
 5497     
 5498     Change-Id: I3e1ec5b1430d5fddee46883d468ff4f5bd6ca54b
 5499     updates: bz#1193929
 5500     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 5501 
 5502 commit b9181e9212b6916ba1c8a269c1474e7aef144f6a
 5503 Author: Yaniv Kaul <ykaul@redhat.com>
 5504 Date:   Mon Nov 4 15:25:27 2019 +0200
 5505 
 5506     socket.c: minor changes
 5507     
 5508     1. Remove dead code and declarations
 5509     2. Move some dict functions to use more efficient ones.
 5510     3. Use more constants, where possible.
 5511     4. Align messages - easier to grep the code for them.
 5512     5. Aligned structures and adding padding where needed.
 5513     
 5514     Change-Id: Ifc2639afe65a935fab5238d3e4a121b662836d3d
 5515     updates: bz#1193929
 5516     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 5517 
 5518 commit 9d0a69187fe085c45b99d8f63febbab0206437c3
 5519 Author: Amar Tumballi <amarts@gmail.com>
 5520 Date:   Thu Oct 17 12:39:02 2019 +0530
 5521 
 5522     libglusterfs: remove unused gfdb specific files from repo
 5523     
 5524     Updates: bz#1193929
 5525     Change-Id: Idb98394c51917e9b132aeb32facccd112effe672
 5526     Signed-off-by: Amar Tumballi <amarts@gmail.com>
 5527 
 5528 commit cd1872ae49d7bc83403fce6821c102edc4e97796
 5529 Author: Mohit Agrawal <moagrawal@redhat.com>
 5530 Date:   Mon Oct 28 21:53:01 2019 +0530
 5531 
 5532     posix: Improve MAKE_HANDLE_PATH
 5533     
 5534     Pass the maximum buffer (PATH_MAX) to posix_handle_path to
 5535     avoid the posix_handle_path call again.
 5536     
 5537     Change-Id: I7d18313870218dc028c5f7fc94d6ec85a2bdb332
 5538     Updates #761
 5539     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 5540 
 5541 commit 515d1b4577d989fa99f236d9e344a71db7d001a0
 5542 Author: Yaniv Kaul <ykaul@redhat.com>
 5543 Date:   Tue Nov 5 18:05:38 2019 +0200
 5544 
 5545     glusterd-volgen.c: improve volgen_graph_set_options_generic()
 5546     
 5547     Skip fetching "skip-CLIOT" unconditionally on every
 5548     invocation of volgen_graph_set_options_generic().
 5549     Instead, fetch only if the vme->key matches to it.
 5550     We calculate the length of vme->key (but we would have
 5551     done it anyway in dict_get() later on, so now we can use
 5552     dict_getn() instead and re-use that key length) and check
 5553     if the lengths match before doing a strcmp() between them.
 5554     Lastly, if they match, we actually do the fetch.
 5555     
 5556     Change-Id: I9d9a7104f9e920bf81477128adb5fc87f5d30627
 5557     updates: bz#1193929
 5558     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 5559 
 5560 commit d1842d97b22419a8b73c2da07caa5f346ee6f7a7
 5561 Author: Amar Tumballi <amarts@gmail.com>
 5562 Date:   Wed Nov 13 11:12:02 2019 +0530
 5563 
 5564     extras: fix the remote subvol name in case of thin-arbiter
 5565     
 5566     * Also make sure to provide log-level in systemd file,
 5567       instead of volfile itself.
 5568     
 5569     Updates: bz#1193929
 5570     Change-Id: I45a0b0ff3ebb4b3e095b4eed14b7bdf5816926d0
 5571     Signed-off-by: Amar Tumballi <amarts@gmail.com>
 5572 
 5573 commit b82b1a468edbfd7a25ed3b99cc8c006a3f798cd1
 5574 Author: N Balachandran <nbalacha@redhat.com>
 5575 Date:   Fri Oct 25 11:25:06 2019 +0530
 5576 
 5577     cluster/dht: Add comments to the code
 5578     
 5579     Add comments to the code to explain what is
 5580     being done and why.
 5581     
 5582     Change-Id: I50831d7bd4bb73e75f6cda05fafaeb5a8619baae
 5583     Updates: bz#1765421
 5584     Signed-off-by: N Balachandran <nbalacha@redhat.com>
 5585 
 5586 commit e97f01ac5c9e2774510807728e52e8acc9524a88
 5587 Author: Ravishankar N <ravishankar@redhat.com>
 5588 Date:   Wed Nov 13 15:39:59 2019 +0530
 5589 
 5590     afr: fix log flooding
 5591     
 5592     Commit "ccf33e789 - dict.c: remove redundant checks"
 5593     removed some NULL checks in certain dict functions. This caused
 5594     flooding of fuse mount logs when I/O was done on the mount on a replica
 5595     volume:
 5596     
 5597     Message:
 5598     W [dict.c:1478:dict_get_with_refn]
 5599     (-->/usr/local/lib/libglusterfs.so.0(dict_get_uint32+0x4d)
 5600     [0x7ff9121ec963] -->/usr/local/lib/libglusterfs.so.0(dict_get_with_ref+0x90)
 5601     [0x7ff9121eb93f] -->/usr/local/lib/libglusterfs.so.0(+0x229be)
 5602     [0x7ff9121eb9be] ) 0-dict: dict OR key (glusterfs.lk.lkmode) is NULL [Invalid argument]
 5603     
 5604     Fix:
 5605     In the relevant AFR functions, check that dict is not NULL before trying
 5606     to perform operations on it.
 5607     
 5608     See bug description for more details.
 5609     fixes: bz#1772006
 5610     Change-Id: I30c89c0b5d6c80cc86a6047aae70127769412120
 5611     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
 5612 
 5613 commit 1e7fc7e540f3f7ca96e16a837d89405901bf03a2
 5614 Author: Mohit Agrawal <moagrawal@redhat.com>
 5615 Date:   Wed Nov 13 08:23:09 2019 +0530
 5616 
 5617     ssl/test: ssl test case is failing when using specific cipher
 5618     
 5619     Problem: On RHEL-8 ssl test case is failing when trying to
 5620              connect with a peer after using the specific cipher.
 5621     
 5622     Solution: If cipher is not supported by openssl on rhel-8 then
 5623               test case is failed.To avoid the issue validate the
 5624               cipher before connecting with peer.
 5625     
 5626     Change-Id: I96d92d3602cf7fd40337126c8305a3f8925faf9b
 5627     Fixes: bz#1756900
 5628     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 5629 
 5630 commit b6b9a27607132cf1b20c34ba948319b27ebce867
 5631 Author: MOHIT AGRAWAL <moagrawa@redhat.com>
 5632 Date:   Wed Nov 13 12:42:13 2019 +0000
 5633 
 5634     Revert "dict.c: remove redundant checks"
 5635     
 5636     This patch floods of dict_log at the time accessing key from dict if
 5637     dict is not valid so revert the patch.
 5638     
 5639     This reverts commit ccf33e789f2cbd202e21340f684e2140d844a129.
 5640     
 5641     Change-Id: I81fa3b080a2f784254b519044ad284858583ec66
 5642     updates: bz#1193929
 5643 
 5644 commit 9a8f8b057415c362485d747c97d4491e80f65a23
 5645 Author: Kotresh HR <khiremat@redhat.com>
 5646 Date:   Wed Nov 13 13:48:28 2019 +0530
 5647 
 5648     geo-rep: Improve debugging
 5649     
 5650     When gsyncd failed with tracebacks during start, it prints
 5651     only exception object but not the error string. This patch
 5652     adds the error string as well.
 5653     
 5654     Earlier:
 5655     "failed with ImportError"
 5656     
 5657     Now:
 5658     "failed with ImportError: No module named _io."
 5659     
 5660     fixes: bz#1771895
 5661     Change-Id: I0d772a250d4c2010a0c35053aa7b165b71f8434e
 5662     Signed-off-by: Kotresh HR <khiremat@redhat.com>
 5663 
 5664 commit e82cd7654a8c2faf46e1c553c7f8c251dff7002c
 5665 Author: N Balachandran <nbalacha@redhat.com>
 5666 Date:   Mon Nov 11 11:25:03 2019 +0530
 5667 
 5668     cluster/dht: Don't skip entries with invalid stat
 5669     
 5670     Don't strip out entries with invalid stats in
 5671     dht_readdirp_cbk.
 5672     
 5673     Change-Id: I136ab342762d020a3c0f43e51e0090aed2af4120
 5674     Fixes: bz#1769754
 5675     Signed-off-by: N Balachandran <nbalacha@redhat.com>
 5676 
 5677 commit 4702232a1febc5de4ce9156e30552a0da4306eb8
 5678 Author: Niklas Hambüchen <mail@nh2.me>
 5679 Date:   Wed Nov 13 12:19:54 2019 +0530
 5680 
 5681     api: Fix documentation path in README
 5682     
 5683     Ref: #84
 5684     Updates: bz#1193929
 5685     Change-Id: If346dbe421d3e301ae035165b2e01ee0a7176d36
 5686     Signed-off-by: Amar Tumballi <amarts@gmail.com>
 5687 
 5688 commit 74592eeb9d7b4dd266c752aaaf8f79b0a519ec4a
 5689 Author: Yaniv Kaul <ykaul@redhat.com>
 5690 Date:   Wed Nov 6 09:52:56 2019 +0200
 5691 
 5692     [WIP]gluster-volgen.c: remove more of JBR and FDL xlators
 5693     
 5694     the JBR and FDL experimental xlators were apparently removed.
 5695     Removed additional leftovers scattered in the code.
 5696     
 5697     Change-Id: I78b6fa5fd9044dc48cdcb1fb094b8c267c2d1323
 5698     updates: bz#1193929
 5699     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 5700 
 5701 commit 9595ecca3de49fdf37d30b151f5c3e071e0a80d0
 5702 Author: Kotresh HR <khiremat@redhat.com>
 5703 Date:   Tue Nov 12 21:53:20 2019 +0530
 5704 
 5705     geo-rep: Fix py2/py3 compatibility in repce
 5706     
 5707     Geo-rep fails to start on python2 only machine like
 5708     centos6. It fails with "ImportError no module named _io".
 5709     This patch fixes the same.
 5710     
 5711     fixes: bz#1771577
 5712     Change-Id: I8228458a853a230546f9faf29a0e9e0f23b3efec
 5713     Signed-off-by: Kotresh HR <khiremat@redhat.com>
 5714 
 5715 commit 256c5dd228af503ece1aebe1229feac61b55f1d3
 5716 Author: Xi Jinyu <xijinyu@cmss.chinamobile.com>
 5717 Date:   Tue Nov 12 01:37:40 2019 -0500
 5718 
 5719     libglusterfs/dict.c: Fix memory leaks.
 5720     
 5721     
 5722     fixes: bz#1771365
 5723     Change-Id: Id31687c4704cbbfea1084f912ed0420c7d1cc1ec
 5724     Signed-off-by: Xi Jinyu <xijinyu@cmss.chinamobile.com>
 5725 
 5726 commit 9f2ad630c877b7b3fb833c89ad85dc10bcf552c4
 5727 Author: Xie Changlong <xiechanglong@cmss.chinamobile.com>
 5728 Date:   Thu Nov 7 17:35:45 2019 +0800
 5729 
 5730     gluster: check ctx->active
 5731     
 5732     To avoid process "TRANSLATOR INFO" "BARRIER" if graph is not ready,
 5733     also see commit ee630e25.
 5734     
 5735     Updates: bz#1769712
 5736     Signed-off-by: Xie Changlong <xiechanglong@cmss.chinamobile.com>
 5737     Change-Id: Ibd446a35962206d3689667cda7e6712d72e4ec2f
 5738 
 5739 commit 45e8ddf6acebaf8c178b432024ab82288b54288f
 5740 Author: Amar Tumballi <amarts@gmail.com>
 5741 Date:   Fri Oct 25 22:57:04 2019 +0530
 5742 
 5743     MAINTAINERS: revised the maintainers list after 2 years
 5744     
 5745     The last major change to the file happened 2 years 5 months back.
 5746     For the better health of the project, and to motivate the active
 5747     contributors, and keeping the file up-to-date with who is working
 5748     on what, it is critical for us to keep refreshing the file once
 5749     every 2 years (at least).
 5750     
 5751     This is one such effort.
 5752     
 5753     Highlighted changes:
 5754     
 5755     * Moving both Jeff and Vijay from Maintainers to 'Special Thanks'
 5756       section.
 5757     * Moving Amar and Xavi as Maintainers, and Atin as Peer, mainly
 5758       looking at the activities (patches, reviews, merges) across
 5759       the codebase, and also the contributions in discussions.
 5760     * Moving Shyam and Niels out of Peer list highlighting the 6+
 5761       months of changed priorities.
 5762     
 5763     * Changed Xavi's contact from Datalab's to Red Hat's
 5764     * Changed Amar's contact from Red Hat's to his Personal.
 5765     
 5766     * Removed:
 5767       - Block Device (BD)
 5768       - Experimental (RIO / JBR)
 5769       - Gluster Object
 5770       - Gluster Hadoop Plugin
 5771       - Nagios Monitoring
 5772       - GlusterD2 (and renamed glusterd1 to glusterd)
 5773     
 5774     * Marked 'NFS (gluster-NFS)' component as 'Deprecated / Orphan'
 5775     
 5776     * Moved few people, who stopped major contribution, mainly
 5777       because of changing companies, changing projects inside
 5778       their own company etc, to 'Special Thanks' section.
 5779     
 5780     * Additions:
 5781       - Kotresh added as Peer in Posix
 5782       - Raghavendra Gowdappa added as Peer in FUSE Bridge,
 5783         Maintainer of readdir-ahead
 5784       - Nithya Added as peer in readdir-ahead
 5785       - Hari Gowtham as Peer in Quota, Maintainer in Releases
 5786       - Yaniv Kaul for xxhash
 5787       - Rinku Kothiya added as Peer in releases.
 5788       - Sheetal Pamecha and Shwetha Acharya added as Peers in
 5789         packaging for Debian/Ubuntu and OpenSUSE.
 5790     
 5791     Updates: bz#1193929
 5792     Change-Id: I0d6eccfee4306e26cdbc2b94f43ac493e2c25a61
 5793     Signed-off-by: Amar Tumballi <amarts@gmail.com>
 5794 
 5795 commit e3d97f57c8e25e8b44d3c96b09d69336ff6edb4b
 5796 Author: Mohit Agrawal <moagrawal@redhat.com>
 5797 Date:   Wed Nov 6 10:32:04 2019 +0530
 5798 
 5799     glusterd: Client Handling of Elastic Clusters
 5800     
 5801     Configure the list of gluster servers in the key
 5802     GLUSTERD_BRICK_SERVERS at the time of GETSPEC RPC CALL
 5803     and access the value in client side to update volfile
 5804     serve list so that client would be able to connect
 5805     next volfile server in case of current volfile server
 5806     is down
 5807     
 5808     Updates #741
 5809     Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
 5810     
 5811     Change-Id: I23f36ddb92982bb02ffd83937a8bd8a2c97e8104
 5812 
 5813 commit 50b6806bb2697246bdc1b9ac5ef19af61584e010
 5814 Author: Sanju Rakonde <srakonde@redhat.com>
 5815 Date:   Tue Oct 22 15:06:29 2019 +0530
 5816 
 5817     cli: display detailed rebalance info
 5818     
 5819     Problem: When one of the node is down in cluster,
 5820     rebalance status is not displaying detailed
 5821     information.
 5822     
 5823     Cause: In glusterd_volume_rebalance_use_rsp_dict()
 5824     we are aggregating rsp from all the nodes into a
 5825     dictionary and sending it to cli for printing. While
 5826     assigning a index to keys we are considering all the
 5827     peers instead of considering only the peers which are
 5828     up. Because of which, index is not reaching till 1.
 5829     while parsing the rsp cli unable to find status-1
 5830     key in dictionary and going out without printing
 5831     any information.
 5832     
 5833     Solution: The simplest fix for this without much
 5834     code change is to continue to look for other keys
 5835     when status-1 key is not found.
 5836     
 5837     fixes: bz#1764119
 5838     
 5839     Change-Id: I0062839933c9706119eb85416256eade97e976dc
 5840     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
 5841 
 5842 commit 5304eaa662b263791baf0e5a9bd616446a3919ef
 5843 Author: Mohit Agrawal <moagrawal@redhat.com>
 5844 Date:   Mon Oct 28 18:42:41 2019 +0530
 5845 
 5846     posix: Improve MAKE_HANDLE_GFID_PATH and MAKE_HANDLE_RELPATH
 5847     
 5848     Avoid one function call to set the gfid_path in buffer
 5849     
 5850     Change-Id: If9b95801b05c34d262fac9a275492c794d12bf58
 5851     Updates #748
 5852     Signed-off-by: Mohit Agrawal <moagrawa@redhat.com>
 5853 
 5854 commit 54ed71dba174385ab0d8fa415e09262f6250430c
 5855 Author: l17zhou <cynthia.zhou@nokia-sbell.com.cn>
 5856 Date:   Mon Nov 4 08:45:52 2019 +0200
 5857 
 5858     rpc: Cleanup SSL specific data at the time of freeing rpc object
 5859     
 5860     Problem: At the time of cleanup rpc object ssl specific data
 5861              is not freeing so it has become a leak.
 5862     
 5863     Solution: To avoid the leak cleanup ssl specific data at the
 5864               time of cleanup rpc object
 5865     
 5866     Credits: l17zhou <cynthia.zhou@nokia-sbell.com.cn>
 5867     Fixes: bz#1768407
 5868     Change-Id: I37f598673ae2d7a33c75f39eb8843ccc6dffaaf0
 5869 
 5870 commit 35d13c1e55e90f8da7d6edaeeab331a66aaf6bfd
 5871 Author: Yaniv Kaul <ykaul@redhat.com>
 5872 Date:   Fri Sep 27 12:22:23 2019 +0300
 5873 
 5874     glusterfsd-mgmt: unify read and write tests
 5875     
 5876     1. Both read and write tests required writing first. Either just
 5877     writing (write test) or write and then read (read test).
 5878     So the code is now unified.
 5879     2. There's no reason to read zeros from /dev/zero. Just use a
 5880     CALLOC'ed buffer.
 5881     I don't think we should read and write zeros, but I did not change
 5882     the code yet (I think compression and/or dedup will offset results)
 5883     
 5884     It appears neither read-perf nor write-perf were tested, so added
 5885     basic tests for them.
 5886     
 5887     Change-Id: I24b1f249fa0335ed652a8982e99c0687d940230e
 5888     updates: bz#1193929
 5889     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 5890 
 5891 commit ccf33e789f2cbd202e21340f684e2140d844a129
 5892 Author: Yaniv Kaul <ykaul@redhat.com>
 5893 Date:   Wed Oct 9 21:27:54 2019 +0300
 5894 
 5895     dict.c: remove redundant checks
 5896     
 5897     It removes ~2% of the object size, and did not add meaningful checks
 5898     whatsoever or to readability (IMHO).
 5899     
 5900     Change-Id: I54346c2af47dcab13bd8d5c8ad52fc765e6afad4
 5901     updates: bz#1193929
 5902     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 5903 
 5904 commit 1de29634e5e74610570fe6f378a994bdf65c34cb
 5905 Author: Yaniv Kaul <ykaul@redhat.com>
 5906 Date:   Wed Oct 2 08:13:57 2019 +0300
 5907 
 5908     iobuf.c, h: minor fixes
 5909     
 5910     - Align structures
 5911     - gf_iobuf_get_pagesize() will now also return the index, reducing the need
 5912     for an additional very similar call.
 5913     - Removal of an inefficient loop I've inadvertently added previously.
 5914     It was harmless, but just inefficient.
 5915     - New pool initialization does not need to be done under lock - no
 5916     one can touch that pool yet, so no need to protect it.
 5917     
 5918     Change-Id: I61c50f2f14fa79edc131e515e9615a9928ee2dca
 5919     updates: bz#1193929
 5920     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 5921 
 5922 commit 6b05f29808b187ce720c26e226be673b917ae668
 5923 Author: Csaba Henk <csaba@redhat.com>
 5924 Date:   Sat Sep 21 00:00:07 2019 +0200
 5925 
 5926     extras: add statedumpparse.rb
 5927     
 5928     By default, the script mangles statedump records
 5929     to JSON, which has two benefits:
 5930     - easier machine processing
 5931     - more friendly with line oriented tools, as one
 5932       line will correspond to one record
 5933     
 5934     '--format=memstat' is also available which displays
 5935     memory allocation types along the size of their allocations.
 5936     
 5937     Change-Id: I1685d3afcea6009fbcfafb33798f85bcd645c82f
 5938     updates: bz#1193929
 5939     Signed-off-by: Csaba Henk <csaba@redhat.com>
 5940 
 5941 commit 5c05bc7bcfc47dfb5ead3c07fa42311bf88091b6
 5942 Author: Csaba Henk <csaba@redhat.com>
 5943 Date:   Thu Jul 25 08:27:24 2019 +0200
 5944 
 5945     Improve logging of fusedump creation failure
 5946     
 5947     updates bz#1193929
 5948     Signed-off-by: Csaba Henk <csaba@redhat.com>
 5949     Change-Id: I12cbe1d87f60fb497654d0e13e12171940867f76
 5950 
 5951 commit 0fc68040b72fc94dec3874345547e294b9ec1f45
 5952 Author: Aravinda VK <avishwan@redhat.com>
 5953 Date:   Wed Oct 23 10:10:12 2019 +0530
 5954 
 5955     georep: Merge Worker and Agent as a single process
 5956     
 5957     - libgfchangelog is simplified by removing unnecessary API Class
 5958     - Merged Agent logic into Worker instead of running Worker and Agent as
 5959       two separate processes and maintaining RPC between Worker and Agent.
 5960     - Geo-rep command Pause and Resume will continue without any changes.
 5961       But Agent functionality also gets paused with that.
 5962     
 5963     Updates: #755
 5964     Change-Id: Ie2c00fa7dddf21f180f0649e0aaf084d29023c98
 5965     Signed-off-by: Aravinda VK <avishwan@redhat.com>
 5966 
 5967 commit 0ab6c178468b6cce095c54ab62cfa51162d01fcc
 5968 Author: Yaniv Kaul <ykaul@redhat.com>
 5969 Date:   Fri Oct 11 01:21:43 2019 +0300
 5970 
 5971     cloudsync: remove dead code, unused variable
 5972     
 5973     Minor changes - remove unused functions and unused variables.
 5974     Switch dict functions to use sizen() when relevant.
 5975     
 5976     Change-Id: I737ce04a4beaf4df9b1eea25a90100d315627c14
 5977     updates: bz#1193929
 5978     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 5979 
 5980 commit b17a65dc2e3b4eca0ddd3caf46120f35ed4574b2
 5981 Author: Yaniv Kaul <ykaul@redhat.com>
 5982 Date:   Sun Oct 20 09:12:20 2019 +0300
 5983 
 5984     posix-helpers.c (and others): minor changes to posix_xattr_fill()
 5985     
 5986     posix_xattr_fill() is called from several POSIX functions.
 5987     Made minor changes to it and the functions called from it:
 5988     1. Dict functions to use known lengths (dict_getn() instead of dict_get(), etc.)
 5989     2. Re-ordered some static char[] arrays, to account (hopefully)
 5990     to the frequency of the xattrs usage (based on grep in the code...)
 5991     3. Before strcmp(), check if the strings lengths match.
 5992     4. Removed some dead code.
 5993     
 5994     Hopefully, no functional changes.
 5995     
 5996     Change-Id: I510c0d2785e54ffe0f82c4c449782f2302d63a32
 5997     updates: bz#1193929
 5998     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 5999 
 6000 commit 6c08e5b0ed2bfdd46064187ba39a30c91913fcd7
 6001 Author: Hari Gowtham <hgowtham@redhat.com>
 6002 Date:   Thu Oct 24 17:40:44 2019 +0530
 6003 
 6004     scripts: quota_fsck script TypeError: %d format:not dict
 6005     
 6006     Problem: One of the prints in the script have been using
 6007     %i as the format for printing which doesn't work.
 6008     
 6009     Fix: use %s as the format in the place of %i
 6010     
 6011     Fixes: bz#1764129
 6012     Change-Id: I4480ede7bf62906ddedbe5f880a1e89c76946641
 6013     Signed-off-by: Hari Gowtham <hgowtham@redhat.com>
 6014 
 6015 commit 93e31716f11cb10d21f6fc59e8cdb6baed50865d
 6016 Author: Purna Pavan Chandra Aekkaladevi <paekkala@redhat.com>
 6017 Date:   Fri Oct 18 12:22:13 2019 +0530
 6018 
 6019     barrier: structs re-alignment
 6020     
 6021     Re-aligned a struct and added manual padding to avoid compile time padding.
 6022     
 6023     Change-Id: I8c6fc67a1b635dc7d8e2522bd4fd8587c09f74e9
 6024     Updates: bz#1754448
 6025     Signed-off-by: Purna Pavan Chandra Aekkaladevi <paekkala@redhat.com>
 6026 
 6027 commit 28199dc009869858bcea108f116ce10c06358e03
 6028 Author: Yaniv Kaul <ykaul@redhat.com>
 6029 Date:   Wed May 22 17:36:11 2019 +0300
 6030 
 6031     dict.c: remove one strlen() done under lock.
 6032     
 6033     It's not needed in some cases, where we can pass the key length.
 6034     
 6035     Change-Id: I2e8ed0b69c50a35f5858bae7c1755a0dfd2c77e5
 6036     updates: bz#1193929
 6037     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 6038 
 6039 commit 3ab23415804502b1ba89360c55ac3e8143822a0b
 6040 Author: Ashish Pandey <aspandey@redhat.com>
 6041 Date:   Thu Jul 11 16:52:49 2019 +0530
 6042 
 6043     cluster/ec: Change handling of heal failure to avoid crash
 6044     
 6045     Problem:
 6046     ec_getxattr_heal_cbk was called with NULL as second argument
 6047     in case heal was failing.
 6048     This function was dereferencing "cookie" argument which caused crash.
 6049     
 6050     Solution:
 6051     Cookie is changed to carry the value that was supposed to be
 6052     stored in fop->data, so even in the case when fop is NULL in error
 6053     case, there won't be any NULL dereference.
 6054     
 6055     Thanks to Xavi for the suggestion about the fix.
 6056     
 6057     Change-Id: I0798000d5cadb17c3c2fbfa1baf77033ffc2bb8c
 6058     fixes: bz#1729085
 6059 
 6060 commit 7a1c912a6b145cbf3b22884a7a6b2dfe16fb5de0
 6061 Author: Yaniv Kaul <ykaul@redhat.com>
 6062 Date:   Thu Oct 31 18:13:16 2019 +0200
 6063 
 6064     posix-entry-ops.c: pass correct key length in posix_skip_non_linkto_unlink()
 6065     
 6066     As was found out in a different patch, passing a pointer
 6067     to a constant string still calculates the size of the pointer
 6068     and not the string. This is not catastrophic as the dictionary
 6069     is matching not just hashes but key names, but is certainly what
 6070     was intended to be done.
 6071     Instead, pass explicitly the key lenth.
 6072     
 6073     I've looked for additional cases for such an issue and did not
 6074     find them, luckily.
 6075     
 6076     Change-Id: I644a07a77dd541e9cf4886811ab54897e0c9d483
 6077     updates: bz#1193929
 6078     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 6079 
 6080 commit 13cb14811cdea477780e58ba35493479c7a04a25
 6081 Author: Sheetal Pamecha <spamecha@redhat.com>
 6082 Date:   Tue Oct 22 14:34:06 2019 +0530
 6083 
 6084     tests/shard: fix tests/bugs/shard/unlinks-and-renames.t failure
 6085     
 6086     on rhel8 machine cleanup of shards is not happening properly for a
 6087     sharded file with hard-links. It needs to refresh the hard link count
 6088     to make it successful
 6089     
 6090     The problem occurs when a sharded file with hard-links gets removed.
 6091     When the last link file is removed, all shards need to be cleaned up.
 6092     But in the current code structure shard xlator, instead of sending a lookup
 6093     to get the link count uses stale cache values of inodectx. Therby removing
 6094     the base shard but not the shards present in /.shard directory.
 6095     
 6096     This fix will make sure that it marks in the first unlink's callback that
 6097     the inode ctx needs a refresh so that in the next operation, it will be
 6098     refreshed by looking up the file on-disk.
 6099     
 6100     fixes: bz#1764110
 6101     Change-Id: I81625c7451dabf006c0864d859b1600f3521b648
 6102     Signed-off-by: Sheetal Pamecha <spamecha@redhat.com>
 6103 
 6104 commit 42c1605f42b89520d4d05806d7074e9e93b63640
 6105 Author: Milind Changire <mchangir@redhat.com>
 6106 Date:   Thu Jan 5 19:53:19 2017 +0530
 6107 
 6108     tools/glusterfind: handle offline bricks
 6109     
 6110     Problem:
 6111     glusterfind is unable to copy remote output file to local node when a
 6112     remove-brick is in progress on the remote node. After copying remote
 6113     files, in the --full output listing path, a "sort -u" command is run on
 6114     the collected files. However, "sort" exits with an error code if it
 6115     finds any file missing.
 6116     
 6117     Solution:
 6118     Maintain a map of (pid, output file) when the node commands are started
 6119     and remove the mapping for the pid for which the command returns an
 6120     error. Use the list of files present in the map for the "sort" command.
 6121     
 6122     Change-Id: Ie6e019037379f4cb163f24b1c65eb382efc2fb3b
 6123     fixes: bz#1410439
 6124     Signed-off-by: Milind Changire <mchangir@redhat.com>
 6125     Signed-off-by: Shwetha K Acharya <sacharya@redhat.com>
 6126 
 6127 commit f306a170e70fe9216dcc8e8bb943cf3704cca7d3
 6128 Author: Kotresh HR <khiremat@redhat.com>
 6129 Date:   Fri Oct 25 17:02:30 2019 +0530
 6130 
 6131     MAINTAINERS: Add Sunny Kumar as co-maintainer of geo-rep
 6132     
 6133     Change-Id: I774710273702f1d093601035435989d3caba449c
 6134     Signed-off-by: Kotresh HR <khiremat@redhat.com>
 6135     fixes: bz#1765542
 6136 
 6137 commit f5a14b7aeef2055511de6a9f98e66ce527571ef6
 6138 Author: Atin Mukherjee <amukherj@redhat.com>
 6139 Date:   Wed Oct 23 09:19:19 2019 +0530
 6140 
 6141     MAINTAINERS: Add Sanju & Mohit as glusterd/cli maintainers
 6142     
 6143     Fixes: bz#1764418
 6144     Change-Id: If14c5ce4527844df5251b4b5a720b617815fb1a1
 6145     Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
 6146 
 6147 commit 593150979399f7f11e580591eab4b032bb0228ac
 6148 Author: Ravishankar N <ravishankar@redhat.com>
 6149 Date:   Wed Oct 16 13:06:29 2019 +0530
 6150 
 6151     afr: lock healing changes
 6152     
 6153     Implements lock healing for gluster-block fencing use case.
 6154     
 6155     If mandatory lock is enabled:
 6156     - Add domain lock/unlock to afr_lk fop.
 6157     - Maintain a list of locks to be healed in afr_private_t.
 6158     - Add lock to the list if afr_lk(F_SETLK or F_SETLKW) was sucessful.
 6159     - Remove it from the list during afr_lk(F_UNLCK).
 6160     - On child_down, mark lock as needing heal on that child. If lock is
 6161     lost on quorum no. of bricks, remove it from the list and mark fd bad.
 6162     - For fds marked as bad, fail the subsequent fd based fops.
 6163     - On parent up, traverse the list and heal the locks IFF the client is
 6164     the lk owner and has quorum. (shd does not heal any locks).
 6165     
 6166     updates: #613
 6167     Change-Id: I03c46ceaea30f5e6236d5ec13f71d843d827f1bc
 6168     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
 6169 
 6170 commit 9813ff934f43d09ecbd60b9f73c624b04b7dcc77
 6171 Author: Pranith Kumar K <pkarampu@redhat.com>
 6172 Date:   Thu Oct 24 17:58:03 2019 +0530
 6173 
 6174     cluster/afr: Take a copy of xattr-req
 6175     
 6176     Afr adds its own xattrs to the req, so it should take a copy of the
 6177     dictionary to prevent parent xlator re-using the modified xattr-req
 6178     to another subvolume
 6179     
 6180     fixes: bz#1765155
 6181     Change-Id: I268e2dbd1b12323135d369e90a22a8bdde2cf7c2
 6182     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 6183 
 6184 commit ae6e11982fb7b6e21e2477dae2cecd9f89f50dfc
 6185 Author: Xie Changlong <xiechanglong@cmss.chinamobile.com>
 6186 Date:   Mon Oct 28 17:43:28 2019 +0800
 6187 
 6188     glusterd/ganesha: fix Coverity CID 1405785
 6189     
 6190     To avoid double free
 6191     
 6192     updates: bz#789278
 6193     
 6194     Change-Id: I15ae54ed696295d4cb015668722e77983b062ccb
 6195     Signed-off-by: Xie Changlong <xiechanglong@cmss.chinamobile.com>
 6196 
 6197 commit b6da637929be0bdcdb00bf38fa4c14a8a09209d3
 6198 Author: Pranith Kumar K <pkarampu@redhat.com>
 6199 Date:   Thu Oct 24 12:24:35 2019 +0530
 6200 
 6201     gf-event: Handle unix volfile-servers
 6202     
 6203     Problem:
 6204     glfsheal program uses unix-socket-based volfile server.
 6205     volfile server will be the path to socket in this case.
 6206     gf_event expects this to be hostname in all cases. So getaddrinfo
 6207     will fail on the unix-socket path, events won't be sent in this case.
 6208     
 6209     Fix:
 6210     In case of unix sockets, default to localhost
 6211     
 6212     fixes: bz#1765017
 6213     Change-Id: I60d27608792c29d83fb82beb5fde5ef4754bece8
 6214     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 6215 
 6216 commit f221ba83e11fced96b5a7b506e692358ed278ca2
 6217 Author: Mohammed Rafi KC <rkavunga@redhat.com>
 6218 Date:   Tue Oct 22 16:02:28 2019 +0530
 6219 
 6220     ssl/test: Change the rsa key length to 2048
 6221     
 6222     On a rhel-8 machine, we need to have a key length
 6223     greater than or eaual to 2048. So changing the values
 6224     to 2048 to pass the test.
 6225     
 6226     Credits: Mohit Agrawal <moagrawal@redhat.com>
 6227     
 6228     Change-Id: I0f21db4d737203d0b2e44e7e61f50ae1279795ad
 6229     Updates: bz#1756900
 6230     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
 6231 
 6232 commit f5811979935ce607391825ac6913a95f588818e3
 6233 Author: Mohit Agrawal <moagrawal@redhat.com>
 6234 Date:   Tue Oct 22 18:52:25 2019 +0530
 6235 
 6236     extras: Cgroup(CPU/Mem) restriction are not working on gluster process
 6237     
 6238     Problem: After Configure the Cgroup(CPU/MEM) limit to a gluster processes
 6239              resource(CPU/MEM) limits are not applicable to the gluster
 6240              processes.Cgroup limits are not applicable because all threads are
 6241              not moved into a newly created cgroup to apply restriction.
 6242     
 6243     Solution: To move a gluster thread to newly created cgroup change the
 6244               condition in script
 6245     
 6246     Change-Id: I8ad81c69200e4ec43a74f6052481551cf835354c
 6247     Fixes: bz#1764208
 6248     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 6249 
 6250 commit bfebfa9f2ec9dfc5dbf4a68c3518f98364ebc461
 6251 Author: Sunny Kumar <sunkumar@redhat.com>
 6252 Date:   Thu Oct 24 14:40:37 2019 +0530
 6253 
 6254     test: fix suspicous non-root geo-rep test failures
 6255     
 6256     Export of env variable is required for ssh-copy-id command.
 6257     
 6258     fixes: bz#1765426
 6259     Change-Id: Icaf7a848cb8f4ae9f887d885a8c5bb71f26633b4
 6260     Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
 6261 
 6262 commit 47465050c43f93294a3be27b4b0b1a3715a2c83d
 6263 Author: Vishal Pandey <vpandey@redhat.com>
 6264 Date:   Thu Aug 8 11:46:46 2019 +0530
 6265 
 6266     Worm: xattr update on changing access time of a WORM-Retained file
 6267     
 6268     Retention-period must be updated on changing the access time
 6269     of a worm-retained file. Retention-period must be changed in the
 6270     "trusted.reten-state" xattr
 6271     
 6272     Change-Id: Ieab758a4cf6da3b4bb1d6a3e4f95f400c8a11f1d
 6273     Fixes: bz#1554286
 6274 
 6275 commit 4e9e2cc8f7990ab6deb661c5da98386cd3940d46
 6276 Author: hari gowtham <hgowtham@redhat.com>
 6277 Date:   Tue Oct 22 15:11:03 2019 +0530
 6278 
 6279     Scripts: quota_fsck script KeyError: 'contri_size'
 6280     
 6281     Problem: In a certain code flow, we weren't handling the
 6282     unavailability of the contri value in the dict. Trying to print
 6283     without the value resulted in erroring out.
 6284     
 6285     Fix: Have printed the whole of dictionary as the values will be
 6286     helpful in understanding the state of the file/dir
 6287     
 6288     Fixes: bz#1764129
 6289     Change-Id: I99c538adb712f281ca10e4e0088f404f515b9725
 6290     Signed-off-by: hari gowtham <hgowtham@redhat.com>
 6291 
 6292 commit faf5ac13c4ee00a05e9451bf8da3be2a9043bbf2
 6293 Author: Mohit Agrawal <moagrawal@redhat.com>
 6294 Date:   Thu Oct 3 14:06:52 2019 +0530
 6295 
 6296     rpc: Synchronize slot allocation code
 6297     
 6298     Problem: Current slot allocation/deallocation code path is not
 6299              synchronized.There are scenario when due to race condition
 6300              in slot allocation/deallocation code path brick is crashed.
 6301     
 6302     Solution: Synchronize slot allocation/deallocation code path to
 6303               avoid the issue
 6304     
 6305     Change-Id: I4fb659a75234218ffa0e5e0bf9308f669f75fc25
 6306     Fixes: bz#1763036
 6307     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 6308 
 6309 commit fb79786af4b7b6603a084c430b4542f806716dae
 6310 Author: Kotresh HR <khiremat@redhat.com>
 6311 Date:   Sun Oct 20 01:01:39 2019 +0530
 6312 
 6313     geo-rep: Fix Permission denied traceback on non root setup
 6314     
 6315     Problem:
 6316     While syncing rename of directory in hybrid crawl, geo-rep
 6317     crashes as below.
 6318     
 6319     Traceback (most recent call last):
 6320       File "/usr/local/libexec/glusterfs/python/syncdaemon/repce.py", line 118, in worker
 6321         res = getattr(self.obj, rmeth)(*in_data[2:])
 6322       File "/usr/local/libexec/glusterfs/python/syncdaemon/resource.py", line 588, in entry_ops
 6323         src_entry = get_slv_dir_path(slv_host, slv_volume, gfid)
 6324       File "/usr/local/libexec/glusterfs/python/syncdaemon/syncdutils.py", line 687, in get_slv_dir_path
 6325         [ENOENT], [ESTALE])
 6326       File "/usr/local/libexec/glusterfs/python/syncdaemon/syncdutils.py", line 546, in errno_wrap
 6327         return call(*arg)
 6328     PermissionError: [Errno 13] Permission denied: '/bricks/brick1/b1/.glusterfs/8e/c0/8ec0fcd4-d50f-4a6e-b473-a7943ab66640'
 6329     
 6330     Cause:
 6331     Conversion of gfid to path for a directory uses readlink on backend
 6332     .glusterfs gfid path. But this fails for non root user with
 6333     permission denied.
 6334     
 6335     Fix:
 6336     Use gfid2path interface to get the path from gfid
 6337     
 6338     Change-Id: I9d40c713a1b32cea95144cbc0f384ada82972222
 6339     fixes: bz#1763439
 6340     Signed-off-by: Kotresh HR <khiremat@redhat.com>
 6341 
 6342 commit 61ba30df8c48586ffac957fa312ef7d244b6cdf0
 6343 Author: Yaniv Kaul <ykaul@redhat.com>
 6344 Date:   Fri Oct 18 20:18:21 2019 +0300
 6345 
 6346     posix-entry-ops: do not copy when reading xattr
 6347     
 6348     The code is simplified to avoid needless copy as well as simplified
 6349     overall for readability.
 6350     Such changes are needed elsewhere too (see
 6351     https://github.com/gluster/glusterfs/issues/720 )
 6352     
 6353     Few other minor changes here and there, nothing functional.
 6354     
 6355     updates: bz#1193929
 6356     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 6357     
 6358     Change-Id: I14f9dd2c32a8932bfcc80ebe92c9aa77701095ff
 6359 
 6360 commit d923d3a67d2b306114dec993da7ee9c59d7d5482
 6361 Author: Yaniv Kaul <ykaul@redhat.com>
 6362 Date:   Fri Oct 18 10:46:14 2019 +0300
 6363 
 6364     posix-helper.c: fix compile warn: ‘timeout’ may be used uninitialized
 6365     
 6366     https://review.gluster.org/#/c/glusterfs/+/23439/ introduced this warning:
 6367     In file included from posix-helpers.c:50:
 6368     posix-helpers.c: In function ‘posix_health_check_thread_proc’:
 6369     ../../../../libglusterfs/src/glusterfs/events.h:31:9: warning: ‘timeout’ may be used uninitialized in this function [-Wmaybe-uninitialized]
 6370        31 |         _gf_event(event, ##fmt);                                               \
 6371           |         ^~~~~~~~~
 6372     posix-helpers.c:2024:9: note: ‘timeout’ was declared here
 6373      2024 |     int timeout;
 6374     
 6375     This patch fixes it, by re-setting the initial value to 0.
 6376     
 6377     Change-Id: I90aee4a1366e21a00f5e138f44247d8c42b0f5ae
 6378     updates: bz#789278
 6379     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 6380 
 6381 commit 105c7c5e7deaf9be29b81dc51f58dd5e5f8bd071
 6382 Author: Sanju Rakonde <srakonde@redhat.com>
 6383 Date:   Wed Oct 16 23:26:03 2019 +0530
 6384 
 6385     glusterd: display correct rebalance data size after glusterd restart
 6386     
 6387     Problem: After completion of rebalance, if glusterd is restarted,
 6388     rebalance status displays wrong rebalance data size in its output.
 6389     
 6390     Cause: While glusterd restoring the information from /var/lib/glusterd/
 6391     into its memory, glusterd fetches rebalance_data from
 6392     /var/lib/glusterd/vols/volname/node_state.info. This value is
 6393     converted into an integer using atoi(), which is returning
 6394     incorrect value for larger values.
 6395     
 6396     Solution: use sscanf() instead of atoi() to convert string to
 6397     integer(in this case it is unsigned long)
 6398     
 6399     fixes: bz#1762438
 6400     
 6401     Change-Id: Icbdb096919612b4a1d6fb0e315f09d38900abf4e
 6402     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
 6403 
 6404 commit b0ea2d10f7668f1d1076e06ec165230526d8b892
 6405 Author: Shwetha K Acharya <sacharya@redhat.com>
 6406 Date:   Wed Oct 16 14:53:22 2019 +0530
 6407 
 6408     build: var data is saved under $prefix instead of localstatedir
 6409     
 6410     Problem:glusterfind saves var data under $prefix,
 6411     even though the localstatedir is mentioned while installing i.e, when we
 6412     install glusterfs with ./configure --prefix=/path/to/prefix
 6413     --localstatedir=/path/to/var
 6414     
 6415     This happens because, glusterfind stores working dir as,
 6416     working_dir=@GLUSTERFSD_MISCDIR@/glusterfind/
 6417     and in configure.ac, GLUSTERFSD_MISCDIR is defined as:
 6418     GLUSTERFSD_MISCDIR="$(eval echo $prefix)/var/lib/misc/glusterfsd"
 6419     
 6420     Solution: $localstatedir is used instead of $prefix while defining,
 6421     GLUSTERFSD_MISCDIR to avoid such anomaly.
 6422     
 6423     fixes: bz#1489610
 6424     Change-Id: I431a63c14bee23a1aac612a3dcb0431cb29ca4a2
 6425     Signed-off-by: Shwetha K Acharya <sacharya@redhat.com>
 6426 
 6427 commit f737a745b54c2dffe5e3e3203cea1eecf1c7ecf1
 6428 Author: Yaniv Kaul <ykaul@redhat.com>
 6429 Date:   Fri Sep 13 21:27:44 2019 +0300
 6430 
 6431     rpc: align structs
 6432     
 6433     squash tens of warnings on padding of structs in afr structures.
 6434     The warnings were found by manually added '-Wpadded' to the GCC
 6435     command line.
 6436     
 6437     Also made relevant structs and definitions static, where it
 6438     was applicable.
 6439     
 6440     Change-Id: Ib71a7e9c6179378f072d796d11172d086c343e53
 6441     updates: bz#1193929
 6442     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 6443 
 6444 commit 739940667f4d7f32bd676ba1bea6b1f13426ae03
 6445 Author: Kotresh HR <khiremat@redhat.com>
 6446 Date:   Wed Oct 16 14:25:47 2019 +0530
 6447 
 6448     geo-rep: Fix config upgrade on non-participating node
 6449     
 6450     After upgrade, if the config files are of old format, it
 6451     gets migrated to new format. Monitor process migrates it.
 6452     Since monitor doesn't run on nodes where bricks are not
 6453     hosted, it doesn't get migrated there. So this patch fixes
 6454     the config upgrade on nodes which doesn't host bricks.
 6455     This happens during config either on get/set/reset.
 6456     
 6457     Change-Id: Ibade2f2310b0f3affea21a3baa1ae0eb71162cba
 6458     Signed-off-by: Kotresh HR <khiremat@redhat.com>
 6459     fixes: bz#1762220
 6460 
 6461 commit 1695ece5d70305b4da03d21c3f721ce6d879e4d4
 6462 Author: Ravishankar N <ravishankar@redhat.com>
 6463 Date:   Thu Oct 10 21:59:59 2019 +0530
 6464 
 6465     tests: add tests for  bz#1758878
 6466     
 6467     updates: bz#1193929
 6468     
 6469     Change-Id: I517fa29e57bde970c2c22ebc2de80fec1509cd2d
 6470     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
 6471     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
 6472 
 6473 commit 1cb05730763d98f65c28246c8808217e6e9d6948
 6474 Author: Pranith Kumar K <pkarampu@redhat.com>
 6475 Date:   Mon Oct 14 10:29:31 2019 +0530
 6476 
 6477     tests: Fix spurious failure
 6478     
 6479     fixes: bz#1759002
 6480     Change-Id: I4d49e1c2ca9b3c1d74b9dd5a30f1c66983a76529
 6481     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 6482 
 6483 commit 9e7f5346910b337f87c10e7ae39eecf0bd829e01
 6484 Author: Pranith Kumar K <pkarampu@redhat.com>
 6485 Date:   Tue Oct 15 16:11:00 2019 +0530
 6486 
 6487     tests: Specify bs for dd
 6488     
 6489     On some distros default bs is very slow and the test takes
 6490     close to 2 minutes instead of 20 seconds.
 6491     
 6492     fixes: bz#1761769
 6493     Change-Id: If10d595a7ca05f053237f3c5ffbb09c5151eab35
 6494     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 6495 
 6496 commit 06ed7dc11166975b80c35ec56d82a08b8686f6de
 6497 Author: Pranith Kumar K <pkarampu@redhat.com>
 6498 Date:   Tue Oct 15 15:29:48 2019 +0530
 6499 
 6500     tests: Remove hard-coding of lower bound
 6501     
 6502     Problem:
 6503     Total blocks of XFS partition are dependent on XFS parameters used
 6504     for formatting. So hardcoded lower-bound may not be the lower bound
 6505     for different default parameters.
 6506     
 6507     Used blocks are lesser on my machine for a freshly formatted XFS partition
 6508     compared to where the test fails.
 6509     On my machine:
 6510     Filesystem     1024-blocks  Used Available Capacity Mounted on
 6511     /dev/loop0           98980  5472     93508       6% /d/backends/patchy1
 6512     /dev/loop1           98980  5472     93508       6% /d/backends/patchy2
 6513     
 6514     On a machine where this test fails:
 6515     Filesystem     1024-blocks  Used Available Capacity Mounted on
 6516     /dev/loop0           96928  6112     90816       7% /d/backends/patchy1
 6517     /dev/loop1           96928  6112     90816       7% /d/backends/patchy2
 6518     
 6519     Fix:
 6520     Make lower bound 2% less than the brick-blocks available
 6521     
 6522     fixes: bz#1761759
 6523     Change-Id: I974d5e75766f7ff44780a2e4c2a19cd5d1d14a79
 6524     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 6525 
 6526 commit 9314a9fbf487614c736cf6c4c1b93078d37bb9df
 6527 Author: Mohit Agrawal <moagrawal@redhat.com>
 6528 Date:   Mon Oct 14 15:42:31 2019 +0530
 6529 
 6530     dht: Rebalance causing IO Error - File descriptor in bad state
 6531     
 6532     Problem : When a file is migrated, dht attempts to re-open all open
 6533               fds on the new cached subvol. Earlier, if dht had not opened the fd,
 6534               the client xlator would be unable to find the remote fd and would
 6535               fall back to using an anon fd for the fop. That behavior changed with
 6536               https://review.gluster.org/#/c/glusterfs/+/15804, causing fops to fail
 6537               with EBADFD if the fd was not available on the cached subvol.
 6538               The client xlator returns EBADFD if the remote fd is not found but
 6539               dht only checks for EBADF before re-opening fds on the new cached subvol.
 6540     
 6541     Solution: Handle EBADFD at dht code path to avoid the issue
 6542     
 6543     Change-Id: I43c51995cdd48d05b12e4b2889c8dbe2bb2a72d8
 6544     Fixes: bz#1758579
 6545 
 6546 commit e3c3a3bc8805930b11e890baa99fc6e752e733f5
 6547 Author: Yaniv Kaul <ykaul@redhat.com>
 6548 Date:   Wed Sep 18 21:14:40 2019 +0300
 6549 
 6550     posix-helpers.c: minor changes to posix_fs_health_check
 6551     
 6552     Do not sprintf the path in every invocation.
 6553     Initialize parameters sooner.
 6554     
 6555     Change-Id: I26bfffc5dab6530937a50eee1e7fc4982e9b2a7c
 6556     updates: bz#1193929
 6557     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 6558 
 6559 commit 03e1c3998f4f3b0ade7a0a84b72693da31a213c8
 6560 Author: Sanju Rakonde <srakonde@redhat.com>
 6561 Date:   Mon Oct 14 18:16:21 2019 +0530
 6562 
 6563     tests: mark tests/bugs/glusterd/quorum-value-check.t as NFS test
 6564     
 6565     Fixes: bz#1665358
 6566     
 6567     Change-Id: Iea000dd839d4e4dbef45941f97ab3725a2aa1726
 6568     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
 6569 
 6570 commit b55667cfd78c0d940a54a02bc8f47509ad43904b
 6571 Author: yatipadia <ypadia@redhat.com>
 6572 Date:   Fri Sep 13 13:52:38 2019 +0530
 6573 
 6574     glusterfsd: fix unused value coverity issues
 6575     
 6576     This patch addresses CID-1398624 and CID-1398631
 6577     
 6578     removed the unused variable brick_name
 6579     
 6580     Change-Id: I4f40bd76cb4c94b28589c2333e29d4623da339d0
 6581     Updates: bz#789278
 6582     Signed-off-by: yati <ypadia@redhat.com>
 6583 
 6584 commit 2f43a1a4b567ec02ec03847b6eaec5ac59e68808
 6585 Author: Pranith Kumar K <pkarampu@redhat.com>
 6586 Date:   Thu Oct 10 09:37:24 2019 +0530
 6587 
 6588     cluster/afr: Add afr_seek to fops table
 6589     
 6590     fixes: bz#1760189
 6591     Change-Id: Iffbf8d6f4c50b8e2de8364658697bdbe96549f5d
 6592     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 6593 
 6594 commit c2279f50178be4408a72c04fbbfd6ccc70c50996
 6595 Author: Vishal Pandey <vpandey@redhat.com>
 6596 Date:   Mon Sep 23 20:12:21 2019 +0530
 6597 
 6598     cli: raise error in cli when replica != 2 while ta vol creation
 6599     
 6600     When creating a thin-arbiter volume, if the replica-count
 6601     provided is other than 2, the cli command
 6602     fails(which is the expected behaviour) but it fails
 6603     while transaction is being committed on localhost.
 6604     This, ideally should be handled in cli itself.
 6605     
 6606     Change-Id: I52ecdaf49a7df85670505e4743cdcf3101c71c9f
 6607     Signed-off-by: Vishal Pandey <vpandey@redhat.com>
 6608     Fixes: bz#1754477
 6609 
 6610 commit cdf34880e6788a396c8f02ba7e0263ea09d3676a
 6611 Author: Yaniv Kaul <ykaul@redhat.com>
 6612 Date:   Mon Sep 30 20:27:27 2019 +0300
 6613 
 6614     glusterd-volgen.c: fix setting the key export-dir
 6615     
 6616     The return value of gf_asprintf() was used as the key length,
 6617     and it was stored into ret.
 6618     ret was re-used before the dict function was called, therefore was invalid
 6619     as the real key length and contained a different value.
 6620     
 6621     That was masked by the fact the key length was only used in key hash, so
 6622     while it was incorrect, it was harmless. The hash was consistent and
 6623     the key length was re-calculated anyway in dict_set_lk().
 6624     
 6625     This patch fixes it, so later on we can use the key length also in
 6626     dict_set_lk() to save another strlen() (sent in a different patch).
 6627     
 6628     However, in the course of this patch I've also decided to reduce the
 6629     copy-pasta in this code path and put all NFS options in an array
 6630     and iterate through them. This makes the code shorter and easier to read.
 6631     (It's also more efficient since once an entry was found, there's no
 6632     additional strcmp()'s as the previous code did).
 6633     
 6634     Change-Id: I968ed50a55f3b1a7ad027c72b06e0fa3788eaa9b
 6635     updates: bz#1193929
 6636     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 6637 
 6638 commit 704961a39ffd019487e36ac6dbd425399cb85cf2
 6639 Author: Yaniv Kaul <ykaul@redhat.com>
 6640 Date:   Fri Oct 4 11:17:44 2019 +0300
 6641 
 6642     Multiple files: make root gfid a static variable
 6643     
 6644     In many places we use it, compare to it, etc. It could be a static variable,
 6645     as it really doesn't change. I think it's better than initializing to 0
 6646     and then doing gfid[15] = 1 or other tricks.
 6647     
 6648     I think there are additional oppportunuties to make more variables static.
 6649     This is an attempt at an easy one.
 6650     
 6651     Change-Id: I7f23a30a94056d8f043645371ab841cbd0f90d19
 6652     updates: bz#1193929
 6653     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 6654 
 6655 commit 240eec91c6ab5818c3fa4a9087ba46bbed0e8d2b
 6656 Author: Yaniv Kaul <ykaul@redhat.com>
 6657 Date:   Fri Sep 13 09:20:33 2019 +0300
 6658 
 6659     afr: align structs
 6660     
 6661     squash >50 warnings on padding of structs in afr structures.
 6662     The warnings were found by manually added '-Wpadded' to the GCC
 6663     command line.
 6664     
 6665     Change-Id: I961fbdeb33715cedf3dd10db8e4f8ef40cd3e867
 6666     updates: bz#1193929
 6667     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 6668 
 6669 commit d003ff64778b4d729cec417bfe8f56140806b07d
 6670 Author: Purna Pavan Chandra Aekkaladevi <paekkala@redhat.com>
 6671 Date:   Mon Sep 23 15:36:16 2019 +0530
 6672 
 6673     glusterd: Re-alignment of structure attributes in glusterd.
 6674     
 6675     Structures defined in few files of glusterd src have been modified into
 6676     memory efficient manner.
 6677     
 6678     Change-Id: Ie712b41cc74b04d5169bfd2402bff3b9c0d8c95b
 6679     Updates: bz#1754448
 6680     Signed-off-by: Purna Pavan Chandra Aekkaladevi <paekkala@redhat.com>
 6681 
 6682 commit 1004f959ca3b3697f2759f929ca4c11537f6c488
 6683 Author: yatipadia <ypadia@redhat.com>
 6684 Date:   Fri Oct 4 18:21:42 2019 +0530
 6685 
 6686     POSIX/xlators: Unused and unchecked value coverity issue
 6687     
 6688     This patch addresses CID-1274094 and CID-1382354
 6689     
 6690     Problem(1): "ret" was assigned a value which was never used,
 6691     it was overwritten by 0 and hence has no use.
 6692     
 6693     Problem(2): function was called without checking the return value.
 6694     
 6695     Solution(1): Removed the assignment and just called the function
 6696     whose value was being written in ret.
 6697     
 6698     Solution(2): There was no need to check for the return value as at
 6699     the end 0 is returned, so typecasted the return value as void.
 6700     
 6701     Change-Id: Iefd0e9000c466ef2428c754c31370263bf1ca0d0
 6702     updates: bz#789278
 6703 
 6704 commit 332bac01dd15aaad53ba144c80edca551fe58e11
 6705 Author: Purna Pavan Chandra Aekkaladevi <paekkala@redhat.com>
 6706 Date:   Thu Sep 12 19:50:35 2019 +0530
 6707 
 6708     Posix: UNUSED VALUE coverity fix
 6709     
 6710     This patch fixes Coverity issue with CID 1274206
 6711     
 6712     Problem : -1 is assigned to op_ret, but that stored value is overwritten before it can be used.
 6713     
 6714     Fix : Removal of the line that assigns -1 to op_ret which has no significance.
 6715     
 6716     Change-Id: Icb881549ac946003710551c9b9e88b33b6a06239
 6717     Updates: bz#789278
 6718     Signed-off-by: Purna Pavan Chandra Aekkaladevi <paekkala@redhat.com>
 6719 
 6720 commit 9dd57ff963d4d0dde7f3ad0ff230837ff2f4b1df
 6721 Author: Pranith Kumar K <pkarampu@redhat.com>
 6722 Date:   Thu Oct 10 10:50:59 2019 +0530
 6723 
 6724     performance/open-behind: seek fop should open_and_resume
 6725     
 6726     fixes: bz#1760187
 6727     Change-Id: I4c6ad13194d4fc5c7705e35bf9a27fce504b51f9
 6728     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 6729 
 6730 commit 134fcea8e245fd5033151392b6265d0dcc87a900
 6731 Author: Raghavendra G <rgowdapp@redhat.com>
 6732 Date:   Thu Sep 19 20:22:56 2019 +0530
 6733 
 6734     performance/read-ahead: update expected offset before unwinding read response
 6735     
 6736     With the current code there is a window of time between unwinding
 6737     response to a read request and internal offset is updated to account
 6738     the read just done. If new sequential read request comes in this time
 6739     window, it is incorrectly identified as non-sequential read.
 6740     
 6741     Fix is to update the file offset to account for a read request before
 6742     sending back the response to it.
 6743     
 6744     Change-Id: Iff0c59c769e1eb15f262257763026657e2d4785d
 6745     Signed-off-by: Raghavendra G <rgowdapp@redhat.com>
 6746     Fixes: bz#1753843
 6747 
 6748 commit 3016bfa40ce33f89fbfffb564184576f08caf3d7
 6749 Author: Sanju Rakonde <srakonde@redhat.com>
 6750 Date:   Sun Oct 6 19:05:28 2019 +0530
 6751 
 6752     cli: fix distCount value
 6753     
 6754     gluster volume info --xml id displaying wrong distCount
 6755     value. This patch addresses it.
 6756     
 6757     fixes: bz#1758878
 6758     
 6759     Change-Id: I64081597e06018361e6524587b433b0c4b2a0260
 6760     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
 6761 
 6762 commit 9cfaef0960d206a554035000141935cc0fcf39e5
 6763 Author: Yaniv Kaul <ykaul@redhat.com>
 6764 Date:   Wed Sep 11 17:53:39 2019 +0300
 6765 
 6766     client xlator: misc. cleanups
 6767     
 6768     - remove dead code
 6769     - move functions to be static
 6770     - move some code that only needs to be executed under if branch
 6771     - remove some dead assignments and redundant checks.
 6772     
 6773     No functional change, I hope.
 6774     
 6775     Change-Id: I93d952408197ecd2fa91c3f812a73c54242342fa
 6776     updates: bz#1193929
 6777     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 6778 
 6779 commit 0a9ca8162725b2f6b17a4d2e8cc863d27bb8770e
 6780 Author: Yaniv Kaul <ykaul@redhat.com>
 6781 Date:   Wed Oct 9 00:25:42 2019 +0300
 6782 
 6783     dict.{c,h}: remove the lock from the _data struct
 6784     
 6785     I'm not sure why it was there and I did not see any use for it.
 6786     In the hope I did not miss anything, I removed it.
 6787     
 6788     Change-Id: I02fa2e8e2a598b488fddbff4c7168dc4a41929b2
 6789     updates: bz#1193929
 6790     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 6791 
 6792 commit 79f69d9b4f3c4fdf77dc77989447b6f1d5d5c658
 6793 Author: Yaniv Kaul <ykaul@redhat.com>
 6794 Date:   Thu Oct 3 15:20:20 2019 +0300
 6795 
 6796     glusterd-utils.c: attach_brick() - remove dead code
 6797     
 6798     pidfile1 and pidfile2 were not used anywhere.
 6799     Removed the assignment and the variables.
 6800     
 6801     Change-Id: Ic5fe091ba28bb500c370410a63440953048fd0b7
 6802     updates: bz#1193929
 6803     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 6804 
 6805 commit eae6f48859b81af32e1570e0b2d7a9033c74ac86
 6806 Author: Yaniv Kaul <ykaul@redhat.com>
 6807 Date:   Sat Oct 5 00:20:49 2019 +0300
 6808 
 6809     posix-helpers.c: do not copy when reading xattr
 6810     
 6811     The code is simplified to avoid needless copy as well as simplified
 6812     overall for readability.
 6813     Such changes are needed elsewhere too (see
 6814     https://github.com/gluster/glusterfs/issues/720 )
 6815     
 6816     Few other minor changes here and there, nothing functional.
 6817     
 6818     Change-Id: Ia1167849f54d9cacbfe32ddd712dc1699760daf5
 6819     updates: bz#1193929
 6820     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 6821 
 6822 commit 83173908b3e3a02ebb6eb17963f0889c7ecff987
 6823 Author: Sanju Rakonde <srakonde@redhat.com>
 6824 Date:   Thu Oct 10 20:40:49 2019 +0530
 6825 
 6826     glusterd: rebalance start should fail when quorum is not met
 6827     
 6828     rebalance start should not succeed if quorum is not met.
 6829     this patch adds a condition to check whether quorum is met
 6830     in pre-validation stage.
 6831     
 6832     fixes: bz#1760467
 6833     Change-Id: Ic7d0d08f69e4bc6d5e7abae713ec1881531c8ad4
 6834     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
 6835 
 6836 commit f29e2e640849aee362d847c9994e0a29554dc527
 6837 Author: Kaleb S. KEITHLEY <kkeithle@redhat.com>
 6838 Date:   Wed Oct 9 10:26:39 2019 -0400
 6839 
 6840     packaging: remove leftover bd cruft in rpm .spec
 6841     
 6842     $subject
 6843     
 6844     Change-Id: Ifd8e3340ae6868ea1fd2d0b53f02b093cc23f2cb
 6845     Fixes: bz#1193929
 6846     Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com>
 6847 
 6848 commit 31db2360b6076734f9506e17c4d2c93f7b6ee475
 6849 Author: Pranith Kumar K <pkarampu@redhat.com>
 6850 Date:   Mon Oct 7 12:27:01 2019 +0530
 6851 
 6852     Fix spurious failure in bug-1744548-heal-timeout.t
 6853     
 6854     Script was assuming that the heal would have triggered
 6855     by the time test was executed, which may not be the case.
 6856     It can lead to following failures when the race happens:
 6857     
 6858     ...
 6859     18:29:45 not ok  14 [     85/      1] <  26> '[ 331 == 333 ]' -> ''
 6860     ...
 6861     18:29:45 not ok  16 [  10097/      1] <  33> '[ 668 == 666 ]' -> ''
 6862     
 6863     Heal on 3rd brick didn't start completely first time the command was executed.
 6864     So the extra count got added to the next profile info.
 6865     
 6866     Fixed it by depending on cumulative stats and waiting until the count is
 6867     satisfied using EXPECT_WITHIN
 6868     
 6869     fixes: bz#1759002
 6870     Change-Id: I3b410671c902d6b1458a757fa245613cb29d967d
 6871     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 6872 
 6873 commit 8d17a94a65ac968b179afa718321105856610e63
 6874 Author: karthik-us <ksubrahm@redhat.com>
 6875 Date:   Thu Sep 5 16:14:50 2019 +0530
 6876 
 6877     cluster/afr: Heal entries when there is a source & no healed_sinks
 6878     
 6879     Problem:
 6880     In a situation where B1 blames B2, B2 blames B1 and B3 doesn't blame
 6881     anything for entry heal, heal will not complete even though we have
 6882     clear source and sinks. This will happen because while doing
 6883     afr_selfheal_find_direction() only the bricks which are blamed by
 6884     non-accused bricks are considered as sinks. Later in
 6885     __afr_selfheal_entry_finalize_source() when it tries to mark all the
 6886     non-sources as sinks it fails to do so because there won't be any
 6887     healed_sinks marked, no witness present and there will be a source.
 6888     
 6889     Fix:
 6890     If there is a source and no healed_sinks, then reset all the locked
 6891     sources to 0 and healed sinks to 1 to do conservative merge.
 6892     
 6893     Change-Id: If40d8bc95d52a52b2730f55bdcf135109b421548
 6894     Fixes: bz#1749322
 6895     Signed-off-by: karthik-us <ksubrahm@redhat.com>
 6896 
 6897 commit 65764a8910e02ed2564f92141f6ee735c39b8172
 6898 Author: Ravishankar N <ravishankar@redhat.com>
 6899 Date:   Mon Oct 7 16:59:20 2019 +0530
 6900 
 6901     tests: Fix spurious failure in bug-1134691-afr-lookup-metadata-heal.t
 6902     
 6903     Problem:
 6904     The .t was examining the sink brick's iatt value before the launched
 6905     client-side metadata heal got a chance to complete.
 6906     
 6907     Fix:
 6908     Wait for heal completion.
 6909     
 6910     Fixes: bz#1759081
 6911     Change-Id: I4dd4e3a1cccf35fd18e8cdfea6aa76a726a4763b
 6912     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
 6913 
 6914 commit 47dbd753187f69b3835d2e42fdbe7485874c4b3e
 6915 Author: Ravishankar N <ravishankar@redhat.com>
 6916 Date:   Sat Sep 28 08:53:08 2019 +0530
 6917 
 6918     afr: support split-brain CLI for replica 3
 6919     
 6920     Ever since we added quorum checks for lookups in afr via commit
 6921     bd44d59741bb8c0f5d7a62c5b1094179dd0ce8a4, the split-brain resolution
 6922     commands would not work for replica 3 because there would be no
 6923     readables for the lookup fop.
 6924     
 6925     The argument was that split-brains do not occur in replica 3 but we do
 6926     see (data/metadata) split-brain cases once in a while which indicate that there are
 6927     a few bugs/corner cases yet to be discovered and fixed.
 6928     
 6929     Fortunately, commit  8016d51a3bbd410b0b927ed66be50a09574b7982 added
 6930     GF_CLIENT_PID_GLFS_HEALD as the pid for all fops made by glfsheal. If we
 6931     leverage this and allow lookups in afr when pid is GF_CLIENT_PID_GLFS_HEALD,
 6932     split-brain resolution commands will work for replica 3 volumes too.
 6933     
 6934     Likewise, the check is added in shard_lookup as well to permit resolving
 6935     split-brains by specifying "/.shard/shard-file.xx" as the file name
 6936     (which previously used to fail with EPERM).
 6937     
 6938     Change-Id: I3c543dea79caf7cfbc1633e9089cb1cdd2538ba9
 6939     Fixes: bz#1756938
 6940     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
 6941 
 6942 commit 41752a2f78a3a69fb8e3359906b7132457ca9acc
 6943 Author: Krutika Dhananjay <kdhananj@redhat.com>
 6944 Date:   Wed Sep 11 14:57:29 2019 +0530
 6945 
 6946     extras: Enable direct-io options in group virt
 6947     
 6948     Direct-io options are currently enabled in the ovirt-gluster
 6949     installation through cockpit ui whereas all the other optimizations in
 6950     gluster for virt are enabled through the group-virt option using
 6951     'volume-set' command. For the sake of completeness and to avoid any
 6952     confusion, it would be better to set the o-direct gluster options also
 6953     as part of group virt.
 6954     
 6955     Change-Id: Iac4c443c2a8239effeb05a52344e563f1d877916
 6956     Fixes: bz#1758984
 6957     Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
 6958 
 6959 commit b3df5abbb7ab4d0f5015da9e4d5ecea79eed9b98
 6960 Author: Yaniv Kaul <ykaul@redhat.com>
 6961 Date:   Thu Sep 19 20:52:46 2019 +0300
 6962 
 6963     afr: replace afr_frame_return() when possible with direct call
 6964     
 6965     If you are already under lock, just decrement the call count
 6966     directly instead of removing the lock, re-taking the lock
 6967     and decrementing.
 6968     
 6969     Implements https://github.com/gluster/glusterfs/issues/728
 6970     updates: bz#1193929
 6971     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 6972     
 6973     Change-Id: I3fa20b4651fbdb826655c5a03baeed46e99b5487
 6974 
 6975 commit f2404405432cfc7c2369de02f79eee000b9aee13
 6976 Author: yatipadia <ypadia@redhat.com>
 6977 Date:   Mon Sep 16 14:45:12 2019 +0530
 6978 
 6979     xlators: fixes logically dead code.
 6980     
 6981     This patch addresses CID-1124388.
 6982     
 6983     Problem: When we reach the "out" section in ra_priv_dump(), if the condition
 6984     (ret && conf) holds true, then the value of "add_section" will always be true.
 6985     So the condition  (add_section == _gf_false) will be a dead code.
 6986     
 6987     Fix:"add_section" has no use in the whole block and was making part of the
 6988     block as logically dead code and hence, removed it.
 6989     
 6990     Change-Id: Id7e0105fc9a5ca5b2c2d098c665e6e32ecc6b62b
 6991     updates: bz#789278
 6992     Signed-off-by: yatipadia <ypadia@redhat.com>
 6993 
 6994 commit da331e1e5a04650d340509b4d4de8f20aea0712d
 6995 Author: Amar Tumballi <amarts@gmail.com>
 6996 Date:   Tue Oct 1 15:49:18 2019 +0530
 6997 
 6998     tests: add a pending test case
 6999     
 7000     While merging the protocol handshake fixes (bz#1620580), there was
 7001     a case which was left out. Adding it separately now.
 7002     
 7003     Change-Id: I52133d5fe160b4567400a65e60aac8f7bc20697f
 7004     Updates: bz#1193929
 7005     Signed-off-by: Amar Tumballi <amarts@gmail.com>
 7006 
 7007 commit 9b15867070b0cc241ab165886292ecffc3bc0aed
 7008 Author: N Balachandran <nbalacha@redhat.com>
 7009 Date:   Tue Oct 1 17:37:15 2019 +0530
 7010 
 7011     cluster/dht: Correct fd processing loop
 7012     
 7013     The fd processing loops in the
 7014     dht_migration_complete_check_task and the
 7015     dht_rebalance_inprogress_task functions were unsafe
 7016     and could cause an open to be sent on an already freed
 7017     fd. This has been fixed.
 7018     
 7019     Change-Id: I0a3c7d2fba314089e03dfd704f9dceb134749540
 7020     Fixes: bz#1757399
 7021     Signed-off-by: N Balachandran <nbalacha@redhat.com>
 7022 
 7023 commit 36075d4de78cb562b0e7a8bea85c868251f0e961
 7024 Author: Zhang Huan <zhanghuan@open-fs.com>
 7025 Date:   Wed Jan 24 16:06:08 2018 +0800
 7026 
 7027     rpc: fix missing unref on reconnect
 7028     
 7029     On protocol client connecting to brick, client will firstly contact
 7030     glusterd to get port, then reconnect to glusterfsd. Reconnect cancels
 7031     the reconnect timer and start a new one. However, cancelling the timer
 7032     does not unref rpc ref-ed for it. That leads to refcount leak.
 7033     
 7034     Fix this issue by unref-ing rpc if reconnect timer is canceled.
 7035     
 7036     Change-Id: Ice89dcd93cb283a0c7250c369cc8961d52fb2022
 7037     Fixes: bz#1538900
 7038     BUG: 1538900
 7039     Signed-off-by: Zhang Huan <zhanghuan@open-fs.com>
 7040 
 7041 commit 69aeede2480ff429d1c3c466a79d85fca756b198
 7042 Author: Sanju Rakonde <srakonde@redhat.com>
 7043 Date:   Mon Sep 30 13:14:06 2019 +0530
 7044 
 7045     ssl: fix RHEL8 regression failure
 7046     
 7047     This tests is failing with
 7048     "SSL routines:SSL_CTX_use_certificate:ee key too small"
 7049     in RHEL8. This change is made according to
 7050     https://access.redhat.com/solutions/4157431
 7051     
 7052     updates: bz#1756900
 7053     
 7054     Change-Id: Ib436372c3bd94bcf7324976337add7da4088b3d5
 7055     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
 7056 
 7057 commit 341aab0fbbde0b9867bf8741791afec9354f54d8
 7058 Author: Sanju Rakonde <srakonde@redhat.com>
 7059 Date:   Thu Sep 26 20:07:11 2019 +0530
 7060 
 7061     glusterd: improve logging
 7062     
 7063     updates: bz#1193929
 7064     
 7065     Change-Id: I5b4a39fbdaa43642a322440d550ca24df815cae9
 7066     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
 7067 
 7068 commit 7464ece4b97935b8085ac456a452b697db3bf67e
 7069 Author: Purna Pavan Chandra Aekkaladevi <paekkala@redhat.com>
 7070 Date:   Mon Sep 16 11:05:23 2019 +0530
 7071 
 7072     libgfapi : Dereference after null check - coverity fix
 7073     
 7074     This patch fixes coverity issue with CID 1405581
 7075     
 7076     A pointer 'volname' is being deferenced (by passing it as an argument to strncmp), even when it points to NULL, leading to segmentation fault.
 7077     
 7078     fix : Adding a NULL check condition beforehand and returning NULL
 7079     
 7080     Change-Id: I18685fa1cb251253996051b83b6790a5d5f0aff9
 7081     Updates: bz#789278
 7082     Signed-off-by: Purna Pavan Chandra Aekkaladevi <paekkala@redhat.com>
 7083 
 7084 commit ae729065c0d9cb5411e5c31231b5e293b560d76a
 7085 Author: Amar Tumballi <amarts@gmail.com>
 7086 Date:   Tue Aug 6 23:20:02 2019 +0300
 7087 
 7088     protocol/handshake: pass volume-id for extra check
 7089     
 7090     With added check of volume-id during handshake, we can be sure to not
 7091     connect with a brick if this gets re-used in another volume. This
 7092     prevents any accidental issues which can happen with a stale client
 7093     process lurking along.
 7094     
 7095     Also added test case for testing same volume name which would fetch a
 7096     different volfile (ie, different bricks, different type), and a
 7097     different volume name, but same brick.
 7098     
 7099     For reference:
 7100     Currently a client<->server handshake happens in glusterfs through
 7101     protocol/client translator (setvolume) to protocol/server using a
 7102     dictionary which containes many keys. Rejection happens in server
 7103     side if some of the required keys are missing in handshake
 7104     dictionary.
 7105     
 7106     Till now, there was no single unique identifier to validate for a
 7107     client to tell server if it is actually talking to a corresponding
 7108     server. All we look in protocol/client is a key called
 7109     'remote-subvolume', which should match with a subvolume name in server
 7110     volume file, and for any volume with same brick name (can be present
 7111     in same cluster due to recreate), it would be same. This could cause
 7112     major issue, when a client was connected to a given brick, in one
 7113     volume would be connected to another volume's brick if its
 7114     re-created/re-used.
 7115     
 7116     To prevent this behavior, we are now passing along 'volume-id' in
 7117     handshake, which would be preserved for the life of client process,
 7118     which can prevent this accidental connections.
 7119     
 7120     NOTE: This behavior wouldn't be applicable for user-snapshot enabled
 7121     volumes, as snapshotted volume's would have different volume-id.
 7122     
 7123     Fixes: bz#1620580
 7124     Change-Id: Ie98286e94ce95ae09c2135fd6ec7d7c2ca1e8095
 7125     Signed-off-by: Amar Tumballi <amarts@redhat.com>
 7126 
 7127 commit c6df9e962483bac5bfcd8916318b19040387ce81
 7128 Author: Mohit Agrawal <moagrawal@redhat.com>
 7129 Date:   Thu Sep 26 17:52:30 2019 +0530
 7130 
 7131     posix: heketidbstorage bricks go down during PVC creation
 7132     
 7133     Problem: In OCS environment heketidbstorage is detached due
 7134              to health_check thread is failed.Sometime aio_write
 7135              is not successfully finished within default health-check-timeout
 7136              limit and the brick is detached.
 7137     
 7138     Solution: To avoid the issue increase default timeout to 20s
 7139     
 7140     Change-Id: Idff283d5713da571f9d20a6b296274f69c3e5b7b
 7141     Fixes: bz#1755900
 7142     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 7143 
 7144 commit 257bb3d16b30e60116724e9e906f8d3ecddf313e
 7145 Author: Soumya Koduri <skoduri@redhat.com>
 7146 Date:   Wed Sep 18 16:32:08 2019 +0530
 7147 
 7148     gfapi: Update the version of 'glfs_h_creat_open'
 7149     
 7150     Change-Id: I43eaa53a7b8d876640d225e011b56192bba57eed
 7151     Fixes: bz#1753569
 7152     Signed-off-by: Soumya Koduri <skoduri@redhat.com>
 7153 
 7154 commit 227713229a7bba19426186f1615fd891e4e52bb7
 7155 Author: Pranith Kumar K <pkarampu@redhat.com>
 7156 Date:   Fri Sep 27 10:52:05 2019 +0530
 7157 
 7158     tests/shard: Remove dependence on distributed cache
 7159     
 7160     fixes: bz#1756211
 7161     Change-Id: Iee5b37af89ab624c16a45df364806003238280e5
 7162     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 7163 
 7164 commit 74e810c602a40d6c316f111cc1333c13bd5aec08
 7165 Author: Pranith Kumar K <pkarampu@redhat.com>
 7166 Date:   Tue Sep 10 19:06:54 2019 +0530
 7167 
 7168     cluster/ec: Implement read-mask feature
 7169     
 7170     fixes: #725
 7171     Change-Id: Iaaefe6f49c8193c476b987b92df6bab3e2f62601
 7172     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 7173 
 7174 commit 61f768eddecc52eb1ba44169526dd674864d0aa1
 7175 Author: Xavi Hernandez <jahernan@redhat.com>
 7176 Date:   Wed Sep 25 11:56:35 2019 +0200
 7177 
 7178     cluster/ec: prevent filling shd log with "table not found" messages
 7179     
 7180     When self-heal daemon receives an inodelk contention notification, it tries
 7181     to locate the related inode using inode_find() and the inode table owned by
 7182     top-most xlator, which in this case doesn't have any inode table. This causes
 7183     many messages to be logged by inode_find() function because the inode table
 7184     passed is NULL.
 7185     
 7186     This patch prevents this by making sure the inode table is not NULL before
 7187     calling inode_find().
 7188     
 7189     Change-Id: I8d001bd180aaaf1521ba40a536b097fcf70c991f
 7190     Fixes: bz#1755344
 7191     Signed-off-by: Xavi Hernandez <jahernan@redhat.com>
 7192 
 7193 commit 80cb838773f3574f68b1fa821fcb712ada350309
 7194 Author: Raghavendra Gowdappa <rgowdapp@redhat.com>
 7195 Date:   Tue Feb 12 18:33:44 2019 +0530
 7196 
 7197     read-ahead/io-cache: turn off by default
 7198     
 7199     We've found perf xlators io-cache and read-ahead not adding any
 7200     performance improvement. At best read-ahead is redundant due to kernel
 7201     read-ahead and at worst io-cache is degrading the performance for
 7202     workloads that doesn't involve re-read. Given that VFS already have
 7203     both these functionalities, this patch makes these two
 7204     translators turned off by default for native fuse mounts.
 7205     
 7206     For non-native fuse mounts like gfapi (NFS-ganesha/samba) we can have
 7207     these xlators on by having custom profiles.
 7208     
 7209     Change-Id: Ie7535788909d4c741844473696f001274dc0bb60
 7210     Signed-off-by: Raghavendra Gowdappa <rgowdapp@redhat.com>
 7211     fixes: bz#1676479
 7212 
 7213 commit c4d926900dc36f71c04b3f65ceca5150ce0e8c81
 7214 Author: Mohit Agrawal <moagrawal@redhat.com>
 7215 Date:   Wed Sep 18 19:11:33 2019 +0530
 7216 
 7217     posix: Brick is going down unexpectedly
 7218     
 7219     Problem: In brick_mux environment, while multiple volumes are
 7220              created (1-1000) sometimes brick is going down due to
 7221              health_check thread failure
 7222     
 7223     Solution: Ignore EAGAIN error in health_check thread code to
 7224               avoid the issue
 7225     
 7226     Change-Id: Id44c59f8e071a363a14d09d188813a6633855213
 7227     Fixes: bz#1751907
 7228     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 7229 
 7230 commit 0c2d25956b022cb61f0b62873c97b7346ef7c4c3
 7231 Author: N Balachandran <nbalacha@redhat.com>
 7232 Date:   Wed Sep 25 19:50:27 2019 +0530
 7233 
 7234     perf/write-behind: Clear frame->local on conflict error
 7235     
 7236     WB saves the wb_inode in frame->local for the truncate and
 7237     ftruncate fops. This value is not cleared in case of error
 7238     on a conflicting write request. FRAME_DESTROY finds a non-null
 7239     frame->local and tries to free it using mem_put. However,
 7240     wb_inode is allocated using GF_CALLOC, causing the
 7241     process to crash.
 7242     
 7243     credit: vpolakis@gmail.com
 7244     
 7245     Change-Id: I217f61470445775e05145aebe44c814731c1b8c5
 7246     Fixes: bz#1753592
 7247     Signed-off-by: N Balachandran <nbalacha@redhat.com>
 7248 
 7249 commit 41a0f2aa755ec7162facd30209f2fa3f40308766
 7250 Author: Soumya Koduri <skoduri@redhat.com>
 7251 Date:   Wed Sep 18 16:32:08 2019 +0530
 7252 
 7253     gfapi: 'glfs_h_creat_open' - new API to create handle and open fd
 7254     
 7255     Right now we have two separate APIs, one
 7256     - 'glfs_h_creat_handle' to create handle & another
 7257     - 'glfs_h_open' to create a glfd to return to application
 7258     
 7259     Having two separate routines can result in access errors
 7260     while trying to create and write into a read-only file.
 7261     
 7262     Since a fd is opened even during file/directory creation,
 7263     introducing a new API to make these two operations atomic i.e,
 7264     which can create both handle & fd and pass them to application
 7265     
 7266     Change-Id: Ibf513fcfcdad175f4d7eb6fa7a61b8feec6d33b5
 7267     Fixes: bz#1753569
 7268     Signed-off-by: Soumya Koduri <skoduri@redhat.com>
 7269 
 7270 commit 5677b85a59ec98dbe8c99680308403f56d3ce014
 7271 Author: Sunny Kumar <sunkumar@redhat.com>
 7272 Date:   Tue Sep 24 18:22:13 2019 +0530
 7273 
 7274     tests : test case for non-root geo-rep setup
 7275     
 7276     Added test case for non-root geo-rep setup.
 7277     
 7278     Change-Id: Ib6ebee79949a9f61bdc5c7b5e11b51b262750e98
 7279     fixes: bz#1717827
 7280     Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
 7281 
 7282 commit 74bbad1e73b9b024a666bf5e0fc68465ff0afbdd
 7283 Author: N Balachandran <nbalacha@redhat.com>
 7284 Date:   Fri Sep 20 13:30:42 2019 +0530
 7285 
 7286     glusterfs/fuse: Reduce the default lru-limit value
 7287     
 7288     The current lru-limit value still uses memory for
 7289     upto 128K inodes.
 7290     Reduce the default value of lru-limit to 64K.
 7291     
 7292     Change-Id: Ica2dd4f8f5fde45cb5180d8f02c3d86114ac52b3
 7293     Fixes: bz#1753880
 7294     Signed-off-by: N Balachandran <nbalacha@redhat.com>
 7295 
 7296 commit e15ca15d7cd39d913d9276f93d7ffd2dd7b85611
 7297 Author: Shwetha K Acharya <sacharya@redhat.com>
 7298 Date:   Fri Aug 16 11:37:40 2019 +0530
 7299 
 7300     geo-rep: Fix spelling errors
 7301     
 7302     Fixes: bz#1741779
 7303     Change-Id: I708b6b7e6c520dee10445528e6f99ba38e141c25
 7304     Signed-off-by: Shwetha K Acharya <sacharya@redhat.com>
 7305 
 7306 commit a4900e490c444cf88b202d6925d3ca3a06ad0355
 7307 Author: Sunny Kumar <sunkumar@redhat.com>
 7308 Date:   Sat Sep 21 01:07:30 2019 +0530
 7309 
 7310     geo-rep : fix sub-command during worker connection
 7311     
 7312     Problem:
 7313     
 7314     Geo-rep session for non-root going faulty.
 7315     
 7316     Solution:
 7317     
 7318     During worker start we do not construct slave url and use 'args.resource_remote'
 7319     which is basically just slave-hostname.
 7320     This works better for root session but fails in non-root session during
 7321     ssh command.
 7322     Using slave url solves this issue.
 7323     
 7324     fixes: bz#1753928
 7325     Change-Id: Ib83552fde77f81c208896494b323514ab37ebf22
 7326     Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
 7327 
 7328 commit 9fa05e11b802ddd03825bbe54968e0937878f2b7
 7329 Author: Jiffin Tony Thottan <jthottan@redhat.com>
 7330 Date:   Mon Aug 26 11:32:18 2019 +0530
 7331 
 7332     glusterd/ganesha: fixing resource leak in tear_down_cluster()
 7333     
 7334     CID: 1370947
 7335     Updates: bz#789278
 7336     
 7337     Change-Id: Ib694056430ff0536ed705a0e77e5ace22486891e
 7338     Signed-off-by: Jiffin Tony Thottan <jthottan@redhat.com>
 7339 
 7340 commit 30d3608c43be119f75da7798d88b94601dedcb29
 7341 Author: Sunny Kumar <sunkumar@redhat.com>
 7342 Date:   Fri Sep 20 09:39:12 2019 +0530
 7343 
 7344     geo-rep: performance improvement while syncing renames with existing gfid
 7345     
 7346     Problem:
 7347     The bug[1] addresses issue of data inconsistency when handling RENAME with
 7348     existing destination. This fix requires some performance tuning considering
 7349     this issue occurs in heavy rename workload.
 7350     
 7351     Solution:
 7352     If distribution count for master volume is one do not verify op's on
 7353     master and go ahead with rename.
 7354     
 7355     The performance improvement with this patch can only be observed if
 7356     master volume has distribution count one.
 7357     
 7358     [1]. https://bugzilla.redhat.com/show_bug.cgi?id=1694820
 7359     
 7360     fixes: bz#1753857
 7361     Change-Id: I8e9bcd575e7e35f40f9f78b7961c92dee642f47b
 7362     Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
 7363 
 7364 commit d654dfa526457dc991ebef66fdecda8f0aac529b
 7365 Author: N Balachandran <nbalacha@redhat.com>
 7366 Date:   Fri Sep 20 11:30:45 2019 +0530
 7367 
 7368     mgmt/glusterd: Fixed typos and reworded logs
 7369     
 7370     Fixed typos and reworded log messages for clarity.
 7371     
 7372     Change-Id: I46f616ce7d3eb993c77a5812e8bc044e5f283354
 7373     Fixes: bz#1753859
 7374     Signed-off-by: N Balachandran <nbalacha@redhat.com>
 7375 
 7376 commit c7968f601a7cd4f54103f63d8db1f1a268c0ec5b
 7377 Author: Yaniv Kaul <ykaul@redhat.com>
 7378 Date:   Thu Sep 12 15:19:09 2019 +0300
 7379 
 7380     afr-common.c, afr-self-heal.h: calloc/alloca0 -> malloc/alloca
 7381     
 7382     In 3 cases, there was a memory allocation and zeroing, followed
 7383     directly by populating it with content. Replaced with memory
 7384     allocation that did not zero the memory.
 7385     
 7386     Change-Id: I4fbb5c924fb3a144e415d2368126b784dde760ea
 7387     updates: bz#1193929
 7388     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 7389 
 7390 commit 1088301998d4feabc4285fc6298c08e19a2a558d
 7391 Author: kshithijiyer <kshithij.ki@gmail.com>
 7392 Date:   Wed Jul 10 18:39:51 2019 +0530
 7393 
 7394     [geo-rep] Improving help message for gluster-mountbroker
 7395     
 7396     The arguments for gluster-mountbroker is missing help="< Some Text>"
 7397     parameter wherever add_argument() is used due to which gluster-mountbroker
 7398     help is missing  a brief description of what the argument does in the help
 7399     of {add,setup,remove} as shown below:
 7400     
 7401     usage: gluster-mountbroker remove [-h] [--volume VOLUME] [--user USER]
 7402     
 7403     optional arguments:
 7404       -h, --help       show this help message and exit
 7405       --volume VOLUME
 7406       --user USER
 7407     usage: gluster-mountbroker add [-h] volume user
 7408     
 7409     positional arguments:
 7410       volume
 7411       user
 7412     
 7413     optional arguments:
 7414       -h, --help  show this help message and exit
 7415     usage: gluster-mountbroker setup [-h] mount_root group
 7416     
 7417     positional arguments:
 7418       mount_root
 7419       group
 7420     
 7421     optional arguments:
 7422       -h, --help  show this help message and exit
 7423     
 7424     fixes: bz#1728683
 7425     
 7426     Change-Id: I7eabcd2c103f01e40160ba35500b0a4e5c9f5e7a
 7427     Signed-off-by: kshithijiyer <kshithij.ki@gmail.com>
 7428 
 7429 commit d9ad59c0cdda160aaef6995cce387961ce8939ce
 7430 Author: Kotresh HR <khiremat@redhat.com>
 7431 Date:   Tue Sep 17 15:35:21 2019 +0530
 7432 
 7433     posix/ctime: Fix coverity issue
 7434     
 7435     posix-metadata.c: 462 in posix_set_mdata_xattr()
 7436     ...
 7437     460         GF_VALIDATE_OR_GOTO(this->name, time, out);
 7438     461
 7439     >>>     CID 1405665:  Control flow issues  (DEADCODE)
 7440     >>>     Execution cannot reach the expression "flag->atime" inside this
 7441     >>>     statement: "if (update_utime && (flag->...".
 7442     462         if (update_utime && (flag->ctime && !time) && (flag->atime && !u_atime) &&
 7443     
 7444     Change-Id: Id31d81d04ea2785a669eafe0dc1307303cb2271b
 7445     updates: bz#789278
 7446     Signed-off-by: Kotresh HR <khiremat@redhat.com>
 7447 
 7448 commit 962548166aa68fbf4ac8ef6bdc97657be742e1c5
 7449 Author: N Balachandran <nbalacha@redhat.com>
 7450 Date:   Tue Aug 27 16:35:05 2019 +0530
 7451 
 7452     cluster/dht: Handle file truncates during migration
 7453     
 7454     File truncate operations during a migration were not handled properly.
 7455     This has been fixed.
 7456     
 7457     Change-Id: Ic642d257e893641236a4a21ab69fcc7a569dd70a
 7458     Fixes: bz#1745967
 7459     Signed-off-by: N Balachandran <nbalacha@redhat.com>
 7460 
 7461 commit fc10f370289e32da6753ac9a9a529b60abb16c4d
 7462 Author: Aravinda VK <avishwan@redhat.com>
 7463 Date:   Mon Sep 16 10:04:26 2019 +0530
 7464 
 7465     eventsapi: Set IPv4/IPv6 family based on input IP
 7466     
 7467     server.sin_family was set to AF_INET while creating socket connection,
 7468     this was failing if the input address is IPv6(`::1`).
 7469     
 7470     With this patch, sin_family is set by reading the ai_family of
 7471     `getaddrinfo` result.
 7472     
 7473     Fixes: bz#1752330
 7474     Change-Id: I499f957b432842fa989c698f6e5b25b7016084eb
 7475     Signed-off-by: Aravinda VK <avishwan@redhat.com>
 7476 
 7477 commit 304640e55c0f3c6d15f4e230dc6376e4f5020fea
 7478 Author: Kotresh HR <khiremat@redhat.com>
 7479 Date:   Mon Jul 29 18:30:42 2019 +0530
 7480 
 7481     ctime/rebalance: Heal ctime xattr on directory during rebalance
 7482     
 7483     After add-brick and rebalance, the ctime xattr is not present
 7484     on rebalanced directories on new brick. This patch fixes the
 7485     same.
 7486     
 7487     Note that ctime still doesn't support consistent time across
 7488     distribute sub-volume.
 7489     
 7490     This patch also fixes the in-memory inconsistency of time attributes
 7491     when metadata is self healed.
 7492     
 7493     Change-Id: Ia20506f1839021bf61d4753191e7dc34b31bb2df
 7494     fixes: bz#1734026
 7495     Signed-off-by: Kotresh HR <khiremat@redhat.com>
 7496 
 7497 commit bcd71806b4066b413ee8c3e5a252c1866c9a7e97
 7498 Author: Yaniv Kaul <ykaul@redhat.com>
 7499 Date:   Sat Sep 7 11:28:58 2019 +0300
 7500 
 7501     logging.c: check for log level before checking for args.
 7502     
 7503     In many cases, we will not be logging at all, since the log level
 7504     is irrelevant. In that case, we can just bail out.
 7505     Worth doing this check before the args check, which may not be needed
 7506     in this case.
 7507     
 7508     Change-Id: Ia90a38bb2a49b09bfea5b5abc8b5d0c3bab4e9ac
 7509     updates: bz#1193929
 7510     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 7511 
 7512 commit 0214c32954fcb9384a074c0cfc9e40252a19852e
 7513 Author: Mohammed Rafi KC <rkavunga@redhat.com>
 7514 Date:   Fri Sep 13 15:35:16 2019 +0530
 7515 
 7516     tests/shd: Mark "tests/basic/volume-scale-shd-mux.t" as bad
 7517     
 7518     This test case is failing in upstream. Marking this test as
 7519     bad for now.
 7520     
 7521     Change-Id: I014c67628c14683c32a3c1dd770b10aaf35ad4cc
 7522     Updates: bz#1752331
 7523     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
 7524 
 7525 commit 59f3c790048c9036adb81d56329a66fc3ffb255a
 7526 Author: Raghavendra G <rgowdapp@redhat.com>
 7527 Date:   Mon May 13 12:54:52 2019 +0530
 7528 
 7529     protocol/client: don't reopen fds on which POSIX locks are held after a reconnect
 7530     
 7531     Bricks cleanup any granted locks after a client disconnects and
 7532     currently these locks are not healed after a reconnect. This means
 7533     post reconnect a competing process could be granted a lock even though
 7534     the first process which was granted locks has not unlocked. By not
 7535     re-opening fds, subsequent operations on such fds will fail forcing
 7536     the application to close the current fd and reopen a new one. This way
 7537     we prevent any silent corruption.
 7538     
 7539     A new option "client.strict-locks" is introduced to control this
 7540     behaviour. This option is set to "off" by default.
 7541     
 7542     Change-Id: Ieed545efea466cb5e8f5a36199aa26380c301b9e
 7543     Signed-off-by: Raghavendra G <rgowdapp@redhat.com>
 7544     updates: bz#1694920
 7545 
 7546 commit 61d438e73857776a1f96a7334f56b132275a587b
 7547 Author: Sheetal Pamecha <spamecha@redhat.com>
 7548 Date:   Mon Aug 19 15:27:57 2019 +0530
 7549 
 7550     libgfapi: return correct errno on invalid volume name
 7551     
 7552     glfs_init when called with volume name prefixed by '/'
 7553     sets errno to 0. Setting errno to EINVAL to resolve the issue.
 7554     Also volname is a parameter to glfs_new.
 7555     Thus, validating volname in glfs_new itself and
 7556     returning EINVAL from that function
 7557     
 7558     fixes: bz#1507896
 7559     
 7560     Change-Id: I0d4d2423e26cc07644d50ec8cce788ecc639203d
 7561     Signed-off-by: Sheetal Pamecha <spamecha@redhat.com>
 7562 
 7563 commit da76f7bbef4d311775a5f75c96a54785170a1727
 7564 Author: Soumya Koduri <skoduri@redhat.com>
 7565 Date:   Tue Sep 3 20:52:42 2019 +0530
 7566 
 7567     core/syncop: Bail out if frame creation fails
 7568     
 7569     There could be cases (either due to insufficient memory or
 7570     corrupted mem-pool) due to which frame creation fails. Bail out
 7571     with error in such cases.
 7572     
 7573     Change-Id: I8cc0a5852f6f04d2bac991e4eb79ecb42577da11
 7574     Fixes: bz#1748448
 7575     Signed-off-by: Soumya Koduri <skoduri@redhat.com>
 7576 
 7577 commit 5091a482125eeac3b00915888cf3de775cd3cbbc
 7578 Author: Sheetal <spamecha@redhat.com>
 7579 Date:   Mon Sep 9 22:56:34 2019 +0530
 7580 
 7581     api: Cleanup of executable not done
 7582     
 7583     In test tests/bugs/gfapi/bug-1447266/bug-1447266.t
 7584     actual file created is -
 7585     tests/bugs/gfapi/bug-1447266/bug-1447266
 7586     which is not cleaned up later
 7587     
 7588     fixes: bz#1750618
 7589     Change-Id: I93120418e54b95018a7213d106a1f1c990766281
 7590     Signed-off-by: Sheetal Pamecha <spamecha@redhat.com>
 7591 
 7592 commit 6362999974cc4b55c4c42929b22a2f2f53321699
 7593 Author: Pranith Kumar K <pkarampu@redhat.com>
 7594 Date:   Tue Aug 20 13:27:24 2019 +0530
 7595 
 7596     cluster/ec: Mark release only when it is acquired
 7597     
 7598     Problem:
 7599     Mount-1                                Mount-2
 7600     1)Tries to acquire lock on 'dir1'   1)Tries to acquire lock on 'dir1'
 7601     2)Lock is granted on brick-0        2)Lock gets EAGAIN on brick-0 and
 7602     				      leads to blocking lock on brick-0
 7603     3)Gets a lock-contention            3) Doesn't matter what happens on mount-2
 7604       notification, marks lock->release    from here on.
 7605       to true.
 7606     4)New fop comes on 'dir1' which will
 7607       be put in frozen list as lock->release
 7608       is set to true.
 7609     5) Lock acquisition from step-2 fails because
 7610     3 bricks went down in 4+2 setup.
 7611     
 7612     Fop on mount-1 which is put in frozen list will hang because no codepath will
 7613     move it from frozen list to any other list and the lock will not be retried.
 7614     
 7615     Fix:
 7616     Don't set lock->release to true if lock is not acquired at the time of
 7617     lock-contention-notification
 7618     
 7619     fixes: bz#1743573
 7620     Change-Id: Ie6630db8735ccf372cc54b873a3a3aed7a6082b7
 7621     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 7622 
 7623 commit 880e20ca967d0001e0b6d2780bece8b635a090eb
 7624 Author: Atin Mukherjee <amukherj@redhat.com>
 7625 Date:   Fri Jul 5 09:09:44 2019 +0530
 7626 
 7627     tests: revive back volume-scale-shd-mux.t
 7628     
 7629     Fixes: bz#1708929
 7630     Change-Id: I9cc81a9047ff874df752ca5552e00bf033485bd8
 7631     Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
 7632     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
 7633 
 7634 commit 0e37cdf271a48d3e58c212e95664a2aa34da3940
 7635 Author: Pranith Kumar K <pkarampu@redhat.com>
 7636 Date:   Wed Sep 11 14:41:28 2019 +0530
 7637 
 7638     tests: Fix spurious failure
 7639     
 7640     If heal from next brick starts after the first brick completes heal, then
 7641     opendir on the brick can change atime leading to failure of the test. When
 7642     ctime is disabled it is better to just check mtime to be same after heal.
 7643     
 7644     fixes: bz#1751134
 7645     Change-Id: Ia03e30fd547e6bbe85c1e299845ffa122f3a2692
 7646     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 7647 
 7648 commit 0af88a6c33e8fd57745997afb7c68e4ef6023fbf
 7649 Author: Atin Mukherjee <amukherj@redhat.com>
 7650 Date:   Mon Sep 9 10:01:04 2019 +0530
 7651 
 7652     glusterd, rpc, glusterfsd: fix coverity defects and put required annotations
 7653     
 7654     1404965 - Null pointer dereference
 7655     1404316 - Program hangs
 7656     1401715 - Program hangs
 7657     1401713 - Program hangs
 7658     
 7659     Updates: bz#789278
 7660     Change-Id: I6e6575daafcb067bc910445f82a9d564f43b75a2
 7661     Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
 7662 
 7663 commit ea95631ff47c8048f039faedbc0faa918c4e165a
 7664 Author: Pranith Kumar K <pkarampu@redhat.com>
 7665 Date:   Thu Sep 5 16:12:39 2019 +0530
 7666 
 7667     cluster/ec: quorum-count implementation
 7668     
 7669     fixes: #721
 7670     Change-Id: I5333540e3c635ccf441cf1f4696e4c8986e38ea8
 7671     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 7672 
 7673 commit 6bf09ecbe8203d0f8c020e6b0b55202e91e216c9
 7674 Author: Pranith Kumar K <pkarampu@redhat.com>
 7675 Date:   Sat Sep 7 20:18:01 2019 +0530
 7676 
 7677     cluster/ec: Fix coverity issues
 7678     
 7679     Fixed the following coverity issue in both flush/fsync
 7680     >>>     CID 1404964:  Null pointer dereferences  (REVERSE_INULL)
 7681     >>>     Null-checking "fd" suggests that it may be null, but it has already
 7682     been dereferenced on all paths leading to the check.
 7683     >>>         if (fd != NULL) {
 7684     >>>           fop->fd = fd_ref(fd);
 7685     >>>             if (fop->fd == NULL) {
 7686     >>>                 gf_msg(this->name, GF_LOG_ERROR, 0,
 7687     >>>                        "Failed to reference a "
 7688     >>>                        "file descriptor.");
 7689     
 7690     fixes bz#1748836
 7691     Change-Id: I19c05d585e23f8fbfbc195d1f3775ec528eed671
 7692     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 7693 
 7694 commit 7b3971ad0152eb1bb89a982333970118a6bd4922
 7695 Author: Pranith Kumar K <pkarampu@redhat.com>
 7696 Date:   Wed Sep 4 12:06:34 2019 +0530
 7697 
 7698     cluster/ec: Fail fsync/flush for files on update size/version failure
 7699     
 7700     Problem:
 7701     If update size/version is not successful on the file, updates on the
 7702     same stripe could lead to data corruptions if the earlier un-aligned
 7703     write is not successful on all the bricks. Application won't have
 7704     any knowledge of this because update size/version happens in the
 7705     background.
 7706     
 7707     Fix:
 7708     Fail fsync/flush on fds that are opened before update-size-version
 7709     went bad.
 7710     
 7711     fixes: bz#1748836
 7712     Change-Id: I9d323eddcda703bd27d55f340c4079d76e06e492
 7713     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 7714 
 7715 commit 6563ffb04d7ba51a89726e7c5bbb85c7dbc685b5
 7716 Author: Mohit Agrawal <moagrawal@redhat.com>
 7717 Date:   Mon Sep 2 10:46:10 2019 +0530
 7718 
 7719     glusterd: IPV6 hostname address is not parsed correctly
 7720     
 7721     Problem: IPV6 hostname address is not parsed correctly in function
 7722              glusterd_check_brick_order
 7723     
 7724     Solution: Update the code to parse hostname address
 7725     
 7726     Change-Id: Ifb2f83f9c6e987b2292070e048e97eeb51b728ab
 7727     Fixes: bz#1747746
 7728     Credits: Amgad Saleh <amgad.saleh@nokia.com>
 7729     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 7730 
 7731 commit 66bf17321655225d1096da4f2b5270f02afd102d
 7732 Author: Amar Tumballi <amarts@gmail.com>
 7733 Date:   Tue Sep 3 11:32:39 2019 +0530
 7734 
 7735     rpc/xdr: fixes in Makefile
 7736     
 7737     there is no need to cleanup the .x files.
 7738     
 7739     Fixes: bz#1743094
 7740     Change-Id: I89d8deb3939c83069709c701cb8f1972e3746168
 7741     Signed-off-by: Amar Tumballi <amarts@gmail.com>
 7742 
 7743 commit 43635716e6bd5bd5925fa9194b0853ee919a742d
 7744 Author: Mohammed Rafi KC <rkavunga@redhat.com>
 7745 Date:   Fri Jul 5 20:12:59 2019 +0530
 7746 
 7747     graph/cleanup: Fix race in graph cleanup
 7748     
 7749     We were unconditionally cleaning up the grap when we get
 7750     child_down followed by parent_down. But this is prone to
 7751     race condition when some of the bricks are already disconnected.
 7752     In this case, even before the last child down is executed in the
 7753     client xlator code,we might have freed the graph. Because the
 7754     child_down event is alreadt recevied.
 7755     
 7756     To fix this race, we have introduced a check to see if all client
 7757     xlator have cleared thier reconnect chain, and called the child_down
 7758     for last time.
 7759     
 7760     Change-Id: I7d02813bc366dac733a836e0cd7b14a6fac52042
 7761     fixes: bz#1727329
 7762     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
 7763 
 7764 commit d026f0bcfd301712e4f0671ccf238f43f2e6dd30
 7765 Author: Mohammed Rafi KC <rkavunga@redhat.com>
 7766 Date:   Wed Jul 10 21:44:38 2019 +0530
 7767 
 7768     afr/lookup: Pass xattr_req in while doing a selfheal in lookup
 7769     
 7770     We were not passing xattr_req when doing a name self heal
 7771     as well as a meta data heal. Because of this, some xdata
 7772     was missing which causes i/o errors
 7773     
 7774     Change-Id: Ibfb1205a7eb0195632dc3820116ffbbb8043545f
 7775     Fixes: bz#1728770
 7776     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
 7777 
 7778 commit f6d6c0f65f7b512724694b3436358e1113cbfccb
 7779 Author: Yaniv Kaul <ykaul@redhat.com>
 7780 Date:   Wed Aug 28 19:38:01 2019 +0300
 7781 
 7782     posix*.c: remove unneeded strlen() calls
 7783     
 7784     In various places, we can re-use knowledge of string length
 7785     or result of snprintf() and such instead of strlen().
 7786     
 7787     Change-Id: I4c9b1decf1169b3f8ac83699a0afbd7c38fad746
 7788     updates: bz#1193929
 7789     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 7790 
 7791 commit ee24c0438ce40b0c6faf6581fd4eb162767d8a90
 7792 Author: Yaniv Kaul <ykaul@redhat.com>
 7793 Date:   Fri Aug 30 10:21:33 2019 +0300
 7794 
 7795     glusterd-store.c: remove of dead code
 7796     
 7797     These functions do not seem to be in use.
 7798     
 7799     Change-Id: Ie76baf2a9727b9ba0e66f234226b1e62788245f2
 7800     updates: bz#1193929
 7801     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 7802 
 7803 commit b0dbe11556c17ba6ea0b7d85b741bd4807c81c9c
 7804 Author: kshithijiyer <kshithij.ki@gmail.com>
 7805 Date:   Mon Sep 2 18:43:36 2019 +0530
 7806 
 7807     [Doc] Fixing 404 links in markdown files.
 7808     
 7809     Problem:
 7810     While running markdown-link-checker it was
 7811     observed that there were a large number of
 7812     404 links present in the documentation present
 7813     in the form of markdown files in the project.
 7814     
 7815     This was casued due to the following reasons:
 7816     1. Repos being removed.
 7817     2. Typo in markdown links.
 7818     3. Restructring of directoires.
 7819     
 7820     Solution:
 7821     Fixing all the 404 links present in the project.
 7822     
 7823     fixes: bz#1746810
 7824     Change-Id: I30de745f848fca2e9c92eb7493f74738f0890ed9
 7825     Signed-off-by: kshithijiyer <kshithij.ki@gmail.com>
 7826 
 7827 commit 724c657995a2e148243eeb78c68b620c6d7714a5
 7828 Author: Ravishankar N <ravishankar@redhat.com>
 7829 Date:   Wed Sep 4 11:27:30 2019 +0530
 7830 
 7831     tests: fix spurious failure of bug-1402841.t-mt-dir-scan-race.t
 7832     
 7833     Problem:
 7834     Since commit 600ba94183333c4af9b4a09616690994fd528478, shd starts
 7835     healing as soon as it is toggled from disabled to enabled. This was
 7836     causing the following line in the .t to fail on a 'fast' machine (always
 7837     on my laptop and sometimes on the jenkins slaves).
 7838     
 7839     EXPECT_NOT "^0$" get_pending_heal_count $V0
 7840     
 7841     because by the time shd was disabled, the heal was already completed.
 7842     
 7843     Fix:
 7844     Increase the no. of files to be healed and make it a variable called
 7845     FILE_COUNT, should we need to bump it up further because the machines
 7846     become even faster. Also created pending metadata heals to increase the
 7847     time taken to heal a file.
 7848     
 7849     fixes: bz#1748744
 7850     Change-Id: I5a26b08e45b8c19bce3c01ce67bdcc28ed48198d
 7851     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
 7852 
 7853 commit 80b8cfe3f1386606bada97a76a0cad7acdf6b877
 7854 Author: Mohit Agrawal <moagrawal@redhat.com>
 7855 Date:   Sun Sep 1 12:01:09 2019 +0530
 7856 
 7857     rpc: Update address family if it is not provide in cmd-line arguments
 7858     
 7859     Problem: After enabling transport-type to inet6 and passed ipv6
 7860              transport.socket.bind-address in glusterd.vol clients are
 7861              not started.
 7862     
 7863     Solution: Need to update address-family based on remote-address for
 7864               all gluster client process
 7865     
 7866     Change-Id: Iaa3588cd87cebc45231bfd675745c1a457dc9b31
 7867     Fixes: bz#1747746
 7868     Credits: Amgad Saleh <amgad.saleh@nokia.com>
 7869     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 7870 
 7871 commit a498c478109395e7baa9b029c81d5587254bdd89
 7872 Author: Pranith Kumar K <pkarampu@redhat.com>
 7873 Date:   Wed Aug 28 11:55:17 2019 +0530
 7874 
 7875     debug/error-gen: Set count correctly for short-writes
 7876     
 7877     vector count is reduced to 1 but it is not reflected at the time of
 7878     winding the call. This leads to crash when short-writes are done
 7879     using error-gen. Set the correct count to fix the problem.
 7880     
 7881     fixes: bz#1746320
 7882     Change-Id: I037b60b7e321f2f50b71fb52c43c64707cf114ca
 7883     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 7884 
 7885 commit 7273602b8b88229f182e29a1b977b80b02cb0624
 7886 Author: Jiri Lunacek <jiri.lunacek@hosting90.cz>
 7887 Date:   Tue Aug 6 10:29:34 2019 +0200
 7888 
 7889     peer_map parameter and fix in state detection when no brick is running on peer
 7890     
 7891     The resource agent assumed that peer names in gluster are identical to short hostname of the machine which is not always the case. This commit adds a parameter to solve this.
 7892     Also resource agent reported brick running on cluster nodes that did not cary any brick of specified volume.
 7893     
 7894     Change-Id: I90aacfc34ac0d00aae3f2abb69cacd7278b886bc
 7895     Fixes: bz#1737778
 7896     Signed-off-by: Jiri Lunacek <jiri.lunacek@hosting90.cz>
 7897 
 7898 commit f81bf47d1c3dcb69ec1f2be6f24f2b7fd9737111
 7899 Author: Anoop C S <anoopcs@redhat.com>
 7900 Date:   Thu Aug 29 14:27:12 2019 +0530
 7901 
 7902     build: Fix libglusterd Makefile target
 7903     
 7904     * Fix libglusterd.la target path in cli/src/Makefile.am
 7905     
 7906     * Like libglusterfs, libgfxdr and libgfrpc, libglusterd is also
 7907       expected to be ready by the time xlators/mgmt/glusterd sources
 7908       are compiled. Therefore this change removes the additional
 7909       mentioning of libglusterd.la target in Makefile.am
 7910     
 7911     Change-Id: I1b787316cfb6cd7487f49e661490b9788a0b80b3
 7912     Updates: bz#1193929
 7913     Signed-off-by: Anoop C S <anoopcs@redhat.com>
 7914 
 7915 commit 600ba94183333c4af9b4a09616690994fd528478
 7916 Author: Ravishankar N <ravishankar@redhat.com>
 7917 Date:   Thu Aug 22 17:27:22 2019 +0530
 7918 
 7919     afr: wake up index healer threads
 7920     
 7921     ...whenever shd is re-enabled after disabling or there is a change in
 7922     `cluster.heal-timeout`, without needing to restart shd or waiting for the
 7923     current `cluster.heal-timeout` seconds to expire.
 7924     
 7925     See BZ 1743988 for more details.
 7926     
 7927     Change-Id: Ia5ebd7c8e9f5b54cba3199c141fdd1af2f9b9bfe
 7928     fixes: bz#1744548
 7929     Reported-by: Glen Kiessling <glenk1973@hotmail.com>
 7930     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
 7931 
 7932 commit 77895e92351a8e6257c2e2aaff431446f865fe0b
 7933 Author: Sheetal Pamecha <spamecha@redhat.com>
 7934 Date:   Tue Jul 16 18:43:13 2019 +0530
 7935 
 7936     build: move arch-dependent files from /usr/share to /usr/libexec
 7937     
 7938     All architecture-dependent data should be placed
 7939     in /usr/libexec, and architecture independent data
 7940     files in /usr/share.
 7941     
 7942     fixes: bz#1387404
 7943     Change-Id: Ie3249cae37c5ee4211309ef55a7a76682e57602f
 7944     Signed-off-by: Sheetal Pamecha <spamecha@redhat.com>
 7945 
 7946 commit c90dc63ec9eee0f43ba8e489876fdf8b8810bbdc
 7947 Author: Mohit Agrawal <moagrawal@redhat.com>
 7948 Date:   Wed Aug 28 09:05:20 2019 +0530
 7949 
 7950     glusterd: glusterd service is getting timed out on scaled setup
 7951     
 7952     Problem: On a three node cluster with 2000 replica volumes systemctl is getting
 7953              timed out for glusted service.
 7954     
 7955     Solution: Configure TimeoutSec 300 to wait for glusterd startup.
 7956     
 7957     Change-Id: Idb3f3f3e56e6216a0ebd754cbb9e8e37ce9e636d
 7958     Fixes: bz#1746228
 7959     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 7960 
 7961 commit 5a9ca4b99b256222acf200847c6d4e0447b95652
 7962 Author: N Balachandran <nbalacha@redhat.com>
 7963 Date:   Tue Aug 27 16:57:14 2019 +0530
 7964 
 7965     glusterd: Fixed incorrect size argument
 7966     
 7967     An incorrect size argument to snprintf caused
 7968     the glusterd process to crash on startup. This has been
 7969     fixed.
 7970     
 7971     Change-Id: Iddafb5468866d0182cd8239210c92c893e643285
 7972     Fixes: bz#1745965
 7973     Signed-off-by: N Balachandran <nbalacha@redhat.com>
 7974 
 7975 commit abc472524bc575773ec29c5856f12a29af3af350
 7976 Author: Shwetha K Acharya <sacharya@redhat.com>
 7977 Date:   Fri Jul 12 20:23:20 2019 +0530
 7978 
 7979     build: fix rpmlint warnings in specfile
 7980     
 7981     Fixed following warnings:
 7982     
 7983     glusterfs.spec:286: W: unversioned-explicit-obsoletes %{name}-ufo
 7984     glusterfs.spec:288: W: unversioned-explicit-obsoletes %{name}-rdma
 7985     glusterfs.spec:290: W: unversioned-explicit-obsoletes %{name}-gnfs
 7986     
 7987     fixes: bz#1183054
 7988     Change-Id: I6bc0d3afd34cda1e4a7d74b8006ce3966109638f
 7989     Signed-off-by: Shwetha K Acharya <sacharya@redhat.com>
 7990 
 7991 commit 7926992e65d0a07fdc784a6e45740306d9b4a9f2
 7992 Author: Sanju Rakonde <srakonde@redhat.com>
 7993 Date:   Mon Aug 26 12:38:40 2019 +0530
 7994 
 7995     glusterd: Unused value coverity fix
 7996     
 7997     CID: 1288765
 7998     updates: bz#789278
 7999     
 8000     Change-Id: Ie6b01f81339769f44d82fd7c32ad0ed1a697c69c
 8001     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
 8002 
 8003 commit f0be15cd534b91f88c0239ae850f22c5f2b90c0f
 8004 Author: Atin Mukherjee <amukherj@redhat.com>
 8005 Date:   Mon Jul 15 11:26:55 2019 +0530
 8006 
 8007     glusterd: stop stale bricks during handshaking in brick mux mode
 8008     
 8009     This patch addresses two problems:
 8010     
 8011     1. During friend handshaking, if a volume is imported due to change in
 8012     the version, the old bricks were not stopped which would lead to a
 8013     situation where bricks will run with old volfiles.
 8014     
 8015     2. As part of attaching shd service in glusterd_attach_svc, there might
 8016     be a case that the volume for which we're attempting to attach a shd
 8017     service might become stale and in the process of deletion and hence in
 8018     every retrials (if the rpc connection isn't ready) check for the
 8019     existance of the volume and then only attempt the further attach
 8020     request.
 8021     
 8022     Fixes: bz#1733425
 8023     Change-Id: I6bac6b871f7e31cb5bf277db979289dec196a03e
 8024     Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
 8025     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
 8026 
 8027 commit 093c66a0af496301e5d28d55f4fe57362cb399d5
 8028 Author: Vishal Pandey <vpandey@redhat.com>
 8029 Date:   Wed Aug 7 12:53:06 2019 +0530
 8030 
 8031     glusterd: Add warning and abort in case of failures in migration during remove-brick commit
 8032     
 8033     Problem -
 8034     Currently remove-brick commit goes through even though there were files
 8035     that failed to migrate or were skipped. There is no warning raised to the user.
 8036     Solution-
 8037     Add a check in the remove brick staging phase to verify if the status of the
 8038     rebalnce process is complete but there has been failures or some skipped files
 8039     while migration, In this case user will be given a warning and remove-brick
 8040     commit. User will need to use the force option to remove the bricks.
 8041     
 8042     Fixes: bz#1514683
 8043     Signed-offby- Vishal Pandey <vpandey@redhat.com>
 8044     
 8045     Change-Id: I014d0f0afb4b2fac35ab0de52227f98dbae079d5
 8046 
 8047 commit 30e83561b8dcced2593deb904688177fcd4ff0e0
 8048 Author: Barak Sason <bsasonro@redhat.com>
 8049 Date:   Tue Aug 20 15:37:16 2019 +0300
 8050 
 8051     cli - group files to set volume options supports comments
 8052     
 8053     Modified parsing of group files in such a way that line that starts with
 8054     "#" would be treated as comments
 8055     
 8056     Fixes: bz#1423442
 8057     
 8058     Change-Id: Id85ceb49f8f9c920d4ea551f60bd28767279d4be
 8059     Signed-off-by: Barak Sason <bsasonro@redhat.com>
 8060 
 8061 commit 7afea5eeef61d3f29496b15ba54defeb39161588
 8062 Author: Barak Sason <bsasonro@redhat.com>
 8063 Date:   Sun Aug 18 15:54:11 2019 +0300
 8064 
 8065     cluster/afr - Unused variables
 8066     
 8067     -Minor change to if-else structure to avoid code duplication.
 8068     -Added logging in case method calls fails
 8069     
 8070     CID: 1394654
 8071     Updates: bz#789278
 8072     
 8073     Change-Id: Ibef4450dc89ddd3bf951303d5b87f503924fd250
 8074     Signed-off-by: Barak Sason <bsasonro@redhat.com>
 8075 
 8076 commit 673f2f90f35e5ea381bd22a43403278f89c957a5
 8077 Author: Jiffin Tony Thottan <jthottan@redhat.com>
 8078 Date:   Mon Oct 16 14:24:29 2017 +0530
 8079 
 8080     Revert "glusterd: (storhaug) remove ganesha (843e1b0)"
 8081     
 8082     please note as an additional change, macro GLUSTERD_GET_SNAP_DIR
 8083     moved from glusterd-store.c to glusterd-snapshot-utils.h
 8084     
 8085     Change-Id: I811efefc148453fe32e4f0d322e80455447cec71
 8086     updates: #663
 8087     Signed-off-by: Jiffin Tony Thottan <jthottan@redhat.com>
 8088 
 8089 commit 3fe34c921146a8d11875dc3d904e3a5994eb1cb6
 8090 Author: Jiffin Tony Thottan <jthottan@redhat.com>
 8091 Date:   Mon Oct 16 14:18:31 2017 +0530
 8092 
 8093     Revert "packaging: (ganesha) remove glusterfs-ganesha subpackage and related files"
 8094     
 8095     Until 3.12, glusterd had an option to setup HA cluster for nfs-ganesha using pacemaker
 8096     and corosync. But later infavour of "Storhaug" Project, this functionality was removed
 8097     from the codebase. Since there is not much development happening towards storhaug,
 8098     it better to keep back old working HA solution for nfs-ganesha with bit improvements.
 8099     
 8100     Planned improvements :
 8101     
 8102     * add an option in nfs-ganesha enable to set ganesha without HA.
 8103     
 8104     * Handle usage of export id's properly in the scripts.
 8105     
 8106     Change-Id: I1d60c8970bfc20035cf674d7b2705dfd4819647e
 8107     updates: #663
 8108     Signed-off-by: Jiffin Tony Thottan <jthottan@redhat.com>
 8109 
 8110 commit 032862fa3944fc7152140aaa13cdc474ae594a51
 8111 Author: Mohit Agrawal <moagrawal@redhat.com>
 8112 Date:   Thu Aug 22 15:51:43 2019 +0530
 8113 
 8114     posix: log aio_error return codes in posix_fs_health_check
 8115     
 8116     Problem: Sometime brick is going down to health check thread is
 8117              failed without logging error codes return by aio system calls.
 8118              As per aio_error man page it returns a positive error number
 8119              if the asynchronous I/O operation failed.
 8120     
 8121     Solution: log aio_error return codes in error message
 8122     
 8123     Change-Id: I2496b1bc16e602b0fd3ad53e211de11ec8c641ef
 8124     Fixes: bz#1744519
 8125     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 8126 
 8127 commit 5e52f1900b5c61ea429f25fdcf9dfb6c56f550af
 8128 Author: Kotresh HR <khiremat@redhat.com>
 8129 Date:   Fri Aug 16 16:07:03 2019 +0530
 8130 
 8131     geo-rep: Fix the name of changelog archive file
 8132     
 8133     Background:
 8134     The processed changelogs are archived each month in a single tar file.
 8135     The default format is "archive_YYYYMM.tar" which is specified as "%%Y%%m"
 8136     in configuration file.
 8137     
 8138     Problem:
 8139     The created changelog archive file didn't have corresponding year
 8140     and month. It created as "archive_%Y%m.tar" on python2 only systems.
 8141     
 8142     Cause and Fix:
 8143     Geo-rep expects "%Y%m" after the ConfigParser reads it from config file.
 8144     Since it was "%%Y%%m" in config file, geo-rep used to get correct value
 8145     "%Y%m" in python3 and "%%Y%%m" in python2 which is incorrect.
 8146     The fix can be to use "%Y%m" in config file but that fails in python3.
 8147     So the fix is to use "RawConfigParser" in geo-rep and use "%Y%m". This
 8148     works both in python2 and python3.
 8149     
 8150     Change-Id: Ie5b7d2bc04d0d53cd1769e064c2d67aaf95d557c
 8151     fixes: bz#1741890
 8152     Signed-off-by: Kotresh HR <khiremat@redhat.com>
 8153 
 8154 commit d14d0749340d9cb1ef6fc4b35f2fb3015ed0339d
 8155 Author: Kotresh HR <khiremat@redhat.com>
 8156 Date:   Tue Aug 20 15:49:40 2019 +0530
 8157 
 8158     ctime: Fix incorrect realtime passed to frame->root->ctime
 8159     
 8160     On systems that don't support "timespec_get"(e.g., centos6), it
 8161     was using "clock_gettime" with "CLOCK_MONOTONIC" to get unix epoch
 8162     time which is incorrect. This patch introduces "timespec_now_realtime"
 8163     which uses "clock_gettime" with "CLOCK_REALTIME" which fixes
 8164     the issue.
 8165     
 8166     Change-Id: I57be35ce442d7e05319e82112b687eb4f28d7612
 8167     Signed-off-by: Kotresh HR <khiremat@redhat.com>
 8168     fixes: bz#1743652
 8169 
 8170 commit 8b9b8abd4878d3116dda1c7d7b9736cda09691b6
 8171 Author: Xie Changlong <xiechanglong@cmss.chinamobile.com>
 8172 Date:   Fri Aug 9 14:19:07 2019 +0800
 8173 
 8174     nlm: check if nlm4 is initialized in nlm_priv
 8175     
 8176     Otherwise, gnfs will crash in following steps.
 8177     1) gluster v set <VOL> nfs.disable off
 8178     2) gluster v set <VOL> nfs.nlm off
 8179     3) kill -SIGUSR1 <GNFS_PID>
 8180     4) gnfs crash with SIGSEGV as follows:
 8181     nlm_priv (this=this@entry=0x7f1ad00173b0) at nlm4.c:2742
 8182     0x00007f1acf89d29d in nfs_priv (this=0x7f1ad00173b0) at nfs.c:1662
 8183     0x00007f1ae2941085 in gf_proc_dump_single_xlator_info (trav=trav@entry=0x7f1ad00173b0) at statedump.c:502
 8184     0x00007f1ae29410b8 in gf_proc_dump_per_xlator_info (top=top@entry=0x7f1ad00173b0) at statedump.c:519
 8185     
 8186     fixes: bz#1739360
 8187     Change-Id: Ib9b207a4ccb3226dbc2c449b77de348cbc9a3d3c
 8188     Signed-off-by: Xie Changlong <xiechanglong@cmss.chinamobile.com>
 8189 
 8190 commit 6edaf694e391d91f42a324da6d3a413d66ae74b9
 8191 Author: Barak Sason <bsasonro@redhat.com>
 8192 Date:   Sun Aug 18 17:52:04 2019 +0300
 8193 
 8194     storage/posix - Fixing a coverity issue
 8195     
 8196     Fixed a resource leak of variable 'pfd'
 8197     
 8198     CID: 1400673
 8199     Updates: bz#789278
 8200     
 8201     Change-Id: I78e1e8a89e0604b56e35a75c25d436b35db096c3
 8202     Signed-off-by: Barak Sason <bsasonro@redhat.com>
 8203 
 8204 commit 5ce3ab86a67f99209dd0c533a9bd2a21bd06ef3d
 8205 Author: Barak Sason <bsasonro@redhat.com>
 8206 Date:   Sun Aug 18 11:02:32 2019 +0300
 8207 
 8208     features/utime - fixing a coverity issue
 8209     
 8210     -Modified op_errno init value to a non-negative value in order to avoid
 8211     using a negative value where it's not allowed.
 8212     -In the metod "STACK_UNWIND_STRICT" modified 3rd argument in order to
 8213     represnt the correct value to use (changed from -1 to ret).
 8214     
 8215     CID: 1403650
 8216     Updates: bz#789278
 8217     
 8218     Change-Id: I608031d5af13832e05e180e746b1b5280b54f559
 8219     Signed-off-by: Barak Sason <bsasonro@redhat.com>
 8220 
 8221 commit 4c4eae13b474c32661eeae3734627d1a8494cf54
 8222 Author: Aravinda VK <avishwan@redhat.com>
 8223 Date:   Wed Aug 21 09:54:21 2019 +0530
 8224 
 8225     geo-rep: Structured logging new format
 8226     
 8227     Structured logging format changed in the below patch for better
 8228     readability and parsing. This patch changes the Geo-replication logs
 8229     to that format.
 8230     
 8231     https://review.gluster.org/#/c/glusterfs/+/22987/
 8232     
 8233     Before:
 8234     
 8235     ```
 8236     [2019-08-21 04:23:01.13672] I [monitor(monitor):157:monitor] Monitor: \
 8237       starting gsyncd worker<TAB>brick=/bricks/b1<TAB>slave_node=f29.sonne
 8238     ```
 8239     
 8240     After:
 8241     
 8242     ```
 8243     [2019-08-21 04:23:01.13672] I [monitor(monitor):157:monitor] Monitor: \
 8244       starting gsyncd worker [{brick=/bricks/b1}, {slave_node=f29.sonne}]
 8245     ```
 8246     
 8247     Change-Id: I65ec69a2869e2febeedc558f88620399e4a1a6a4
 8248     Updates: #657
 8249     Signed-off-by: Aravinda VK <avishwan@redhat.com>
 8250 
 8251 commit b960e0103a70b58880b59438b116740af98a4582
 8252 Author: yinkui <13965432176@163.com>
 8253 Date:   Fri Aug 16 10:15:07 2019 +0800
 8254 
 8255     gluster-smb:add smb parameter when  access gluster by cifs
 8256     
 8257     Change-Id: I9ff54f2ca6f86bb5b2f4740485a0159e1fd7785f
 8258     Fixes: bz#1741734
 8259     Signed-off-by: yinkui <13965432176@163.com>
 8260 
 8261 commit fb976636d9ffa3be4d2a47c604a7746be06e2247
 8262 Author: Barak Sason <bsasonro@redhat.com>
 8263 Date:   Thu Aug 15 17:23:17 2019 +0300
 8264 
 8265     features/cloudsync - fix a coverity issue
 8266     
 8267     All assigns to op_errno in this mehod were to the same vlaue - ENOMEM.
 8268     Removed repeted assignments and assigned as init value.
 8269     This also prevents the problem of sending a negatve value of op_errno to
 8270     CS_STACK_UNWIND method
 8271     
 8272     CID: 1394645 - https://scan6.coverity.com/reports.htm#v44018/p10714/fileInstanceId=92065749&defectInstanceId=28018364&mergedDefectId=1394645
 8273     Updates: bz#789278
 8274     
 8275     Change-Id: If765a9216500a38f9392617aaf06583ce36e3262
 8276     Signed-off-by: Barak Sason <bsasonro@redhat.com>
 8277 
 8278 commit 1c3b280602153481fead83622aa885f6735c094c
 8279 Author: Barak Sason <bsasonro@redhat.com>
 8280 Date:   Thu Aug 15 11:06:26 2019 +0300
 8281 
 8282     storage/posix - fixing a coverity issue
 8283     
 8284     CID: 1394644 & 1394639
 8285     Updates: bz#789278
 8286     
 8287     Added logging in case method calls fails
 8288     
 8289     Change-Id: Ib833a5f68d37b98287b84c325637bc688937f647
 8290     Signed-off-by: Barak Sason <bsasonro@redhat.com>
 8291 
 8292 commit 95f71df31dc73d85df722b0e7d3a7eb1e0237e7f
 8293 Author: Kotresh HR <khiremat@redhat.com>
 8294 Date:   Thu Aug 8 10:05:12 2019 +0530
 8295 
 8296     ctime: Fix ctime issue with utime family of syscalls
 8297     
 8298     When atime|mtime is updated via utime family of syscalls,
 8299     ctime is not updated. This patch fixes the same.
 8300     
 8301     Change-Id: I7f86d8f8a1e06a332c3449b5bbdbf128c9690f25
 8302     fixes: bz#1738786
 8303     Signed-off-by: Kotresh HR <khiremat@redhat.com>
 8304 
 8305 commit 72fe2bd9ae0ca223d8a320ca124c2837a1571d82
 8306 Author: Barak Sason <bsasonro@redhat.com>
 8307 Date:   Sun Aug 18 15:36:40 2019 +0300
 8308 
 8309     api: fixing a coverity issue
 8310     
 8311     Added a null pointer check to avoid seg-fault
 8312     
 8313     CID:1403874
 8314     Updates: bz#789278
 8315     
 8316     Change-Id: Ia8ff992c9f06d79647089f296afe77519fb5af2a
 8317     Signed-off-by: Barak Sason <bsasonro@redhat.com>
 8318 
 8319 commit b4b683736367d93daad08a5ee6ca95778c07c5a4
 8320 Author: Anoop C S <anoopcs@redhat.com>
 8321 Date:   Sat Aug 10 10:30:26 2019 +0530
 8322 
 8323     performance/md-cache: Do not skip caching of null character xattr values
 8324     
 8325     Null character string is a valid xattr value in file system. But for
 8326     those xattrs processed by md-cache, it does not update its entries if
 8327     value is null('\0'). This results in ENODATA when those xattrs are
 8328     queried afterwards via getxattr() causing failures in basic operations
 8329     like create, copy etc in a specially configured Samba setup for Mac OS
 8330     clients.
 8331     
 8332     On the other side snapview-server is internally setting empty string("")
 8333     as value for xattrs received as part of listxattr() and are not intended
 8334     to be cached. Therefore we try to maintain that behaviour using an
 8335     additional dictionary key to prevent updation of entries in getxattr()
 8336     and fgetxattr() callbacks in md-cache.
 8337     
 8338     Credits: Poornima G <pgurusid@redhat.com>
 8339     
 8340     Change-Id: I7859cbad0a06ca6d788420c2a495e658699c6ff7
 8341     Fixes: bz#1726205
 8342     Signed-off-by: Anoop C S <anoopcs@redhat.com>
 8343 
 8344 commit faaf22bbe899b0f64b0237398a0d8e3c082f9392
 8345 Author: Kinglong Mee <kinglongmee@gmail.com>
 8346 Date:   Thu Jul 18 11:43:01 2019 +0800
 8347 
 8348     features/locks: avoid use after freed of frame for blocked lock
 8349     
 8350     The fop contains blocked lock may use freed frame info when other
 8351     unlock fop has unwind the blocked lock.
 8352     
 8353     Because the blocked lock is added to block list in inode lock(or
 8354     other lock), after that, when out of the inode lock, the fop
 8355     contains the blocked lock should not use it.
 8356     
 8357     Change-Id: Icb309a1cc78380dc982b26d50c18d67e4f2c8915
 8358     fixes: bz#1737291
 8359     Signed-off-by: Kinglong Mee <mijinlong@horiscale.com>
 8360 
 8361 commit bb01111f50f2585429028c408932416d31c6d316
 8362 Author: Barak Sason <bsasonro@redhat.com>
 8363 Date:   Mon Aug 19 10:34:31 2019 +0300
 8364 
 8365     libglusterfs - fixing a coverity issue
 8366     
 8367     Fixed resource leak of variable volfile_obj
 8368     
 8369     CID: 1402936
 8370     Updates: bz#789278
 8371     
 8372     Change-Id: I1537187e13693686f4ffca891dc1df287fab7a41
 8373     Signed-off-by: Barak Sason <bsasonro@redhat.com>
 8374 
 8375 commit 2fe243d0561c19009964003aa2ced7c695a46d17
 8376 Author: Aravinda VK <avishwan@redhat.com>
 8377 Date:   Wed Jul 3 15:08:31 2019 +0530
 8378 
 8379     logging: Structured logging reference PR
 8380     
 8381     To convert the existing `gf_msg` to `gf_smsg`:
 8382     
 8383     - Define `_STR` of respective Message ID as below(In `*-messages.h`)
 8384     
 8385       #define PC_MSG_REMOTE_OP_FAILED_STR "remote operation failed."
 8386     
 8387     - Change `gf_msg` to use `gf_smsg`. Convert values into fields and
 8388       add any missing fields. Note: `errno` and `error` fields will be
 8389       added automatically to log message in case errnum is specified.
 8390     
 8391       Example:
 8392     
 8393       gf_smsg(
 8394         this->name,                            // Name or log domain
 8395         GF_LOG_WARNING,                        // Log Level
 8396         rsp.op_errno,                          // Error number
 8397         PC_MSG_REMOTE_OP_FAILED,               // Message ID
 8398         "path=%s", local->loc.path,            // Key Value 1
 8399         "gfid=%s", loc_gfid_utoa(&local->loc), // Key Value 2
 8400         NULL                                   // Log End
 8401       );
 8402     
 8403     Key value pairs formatting Help:
 8404     
 8405       gf_slog(
 8406         this->name,                               // Name or log domain
 8407         GF_LOG_WARNING,                           // Log Level
 8408         rsp.op_errno,                             // Error number
 8409         PC_MSG_REMOTE_OP_FAILED,                  // Message ID
 8410         "op=CREATE",                              // Static Key and Value
 8411         "path=%s", local->loc.path,               // Format for Value
 8412         "brick-%d-status=%s", brkidx, brkstatus,  // Use format for key and val
 8413         NULL                                      // Log End
 8414       );
 8415     
 8416     Before:
 8417     
 8418     [2019-07-03 08:16:18.226819] W [MSGID: 114031] [client-rpc-fops_v2.c \
 8419     :2633:client4_0_lookup_cbk] 0-gv3-client-0: remote operation failed. \
 8420     Path: / (00000000-0000-0000-0000-000000000001) [Transport endpoint   \
 8421     is not connected]
 8422     
 8423     After:
 8424     
 8425     [2019-07-29 07:50:15.773765] W [MSGID: 114031]                   \
 8426     [client-rpc-fops_v2.c:2633:client4_0_lookup_cbk] 0-gv1-client-0: \
 8427     remote operation failed. [{path=/f1},                            \
 8428     {gfid=00000000-0000-0000-0000-000000000000},                     \
 8429     {errno=107}, {error=Transport endpoint is not connected}]
 8430     
 8431     To add new `gf_smsg`, Add a Message ID in respective `*-messages.h` file
 8432     and the follow the steps mentioned above.
 8433     
 8434     Change-Id: I4e7d37f27f106ab398e991d931ba2ac7841a44b1
 8435     Updates: #657
 8436     Signed-off-by: Aravinda VK <avishwan@redhat.com>
 8437 
 8438 commit 25cf0d24caa11bdfefe10c93541da07c635d4cfe
 8439 Author: Barak Sason <bsasonro@redhat.com>
 8440 Date:   Sun Aug 18 17:38:09 2019 +0300
 8441 
 8442     mount/fuse - Fixing a coverity issue
 8443     
 8444     Fixed resource leak of dict_value and newkey variables
 8445     
 8446     CID: 1398630
 8447     Updates: bz#789278
 8448     
 8449     Change-Id: I589fdc0aecaeb4f446cd29f95bad36ccd7a35beb
 8450     Signed-off-by: Barak Sason <bsasonro@redhat.com>
 8451 
 8452 commit b1dc7723610325789ba91220ab414e821ec14119
 8453 Author: Kotresh HR <khiremat@redhat.com>
 8454 Date:   Fri Aug 16 15:38:49 2019 +0530
 8455 
 8456     geo-rep: Fix worker connection issue
 8457     
 8458     All the workers connects to primary slave node. It should
 8459     connect to available slave nodes in round robin fashion
 8460     and choose different slave node if the corresponding slave
 8461     node is down. This patch fixes the same.
 8462     
 8463     Thanks Aravinda for the help in root causing this.
 8464     
 8465     Change-Id: I9f8e7744f4adb8a24833cf173681d109710f98cb
 8466     Signed-off-by: Kotresh HR <khiremat@redhat.com>
 8467     Updates: bz#1737484
 8468 
 8469 commit f138d3fa2237e7fa940ecf17153fd700350c4138
 8470 Author: Mohit Agrawal <moagrawal@redhat.com>
 8471 Date:   Tue Jul 16 20:36:57 2019 +0530
 8472 
 8473     posix: In brick_mux brick is crashed while start/stop volume in loop
 8474     
 8475     Problem: In brick_mux environment sometime brick is crashed while
 8476              volume stop/start in a loop.Brick is crashed in janitor task
 8477              at the time of accessing priv.If posix priv is cleaned up before
 8478              call janitor task then janitor task is crashed.
 8479     
 8480     Solution: To avoid the crash in brick_mux environment introduce a new
 8481               flag janitor_task_stop in posix_private and before send CHILD_DOWN event
 8482               wait for update the flag by janitor_task_done
 8483     
 8484     Change-Id: Id9fa5d183a463b2b682774ab5cb9868357d139a4
 8485     fixes: bz#1730409
 8486     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 8487 
 8488 commit a4d073cda22787552b46b186b898304b438db0c2
 8489 Author: Barak Sason <bsasonro@redhat.com>
 8490 Date:   Sun Aug 18 12:21:47 2019 +0300
 8491 
 8492     protocol/client - fixing a coverity issue
 8493     
 8494     Moved null pointer check up in order to avoid seg-fault
 8495     
 8496     CID: 1404258
 8497     Updates: bz#789278
 8498     
 8499     Change-Id: Ib97e05302bfeb8fe38d6ce9870b9740cb576e492
 8500     Signed-off-by: Barak Sason <bsasonro@redhat.com>
 8501 
 8502 commit 0590a1e5dd339af716ed49f0e66f3281ec7a6d0e
 8503 Author: Yaniv Kaul <ykaul@redhat.com>
 8504 Date:   Thu Aug 8 22:23:04 2019 +0300
 8505 
 8506     client_t.c: removal of dead code.
 8507     
 8508     Change-Id: Id9f5f448db305f3135a1fdca61b1d7ec898c63a4
 8509     updates: bz#1193929
 8510     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 8511 
 8512 commit c63a1f6ab0f2ff08e0a738f33558a3f8bdd87996
 8513 Author: Barak Sason <bsasonro@redhat.com>
 8514 Date:   Thu Aug 15 15:54:28 2019 +0300
 8515 
 8516     storage/posix - Moved pointed validity check in order to avoid possible seg-fault
 8517     
 8518     CID: 1124831
 8519     Updates: bz#789278
 8520     
 8521     Change-Id: Ia6550be3742849809cf3e0a4a39d9d6e77003b35
 8522     Signed-off-by: Barak Sason <bsasonro@redhat.com>
 8523 
 8524 commit 8f1620ad7f5d3d040fee55c5f873349800e2268d
 8525 Author: Mohit Agrawal <moagrawal@redhat.com>
 8526 Date:   Mon Aug 19 16:00:24 2019 +0530
 8527 
 8528     glusterd: ./tests/bugs/glusterd/bug-1595320.t is failing
 8529     
 8530     Problem: sometime ./tests/bugs/glusterd/bug-1595320.t is failing  is
 8531              failing at the time of checking brick_process after sending
 8532              a kill signal to brick process
 8533     
 8534     Solution: Wait sometime after just sending a kill signal to brick
 8535               process to make sure brick process is stopped
 8536     
 8537     Change-Id: Iee9e91284618abfc62a550d47e4f9117785def58
 8538     Fixes: bz#1743200
 8539     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 8540 
 8541 commit c2b0696f0c2b901b8c34c732c1ba85c364ff2f04
 8542 Author: N Balachandran <nbalacha@redhat.com>
 8543 Date:   Fri Sep 7 20:48:59 2018 +0530
 8544 
 8545     tests/dht: Add a test file for file renames
 8546     
 8547     Test the various combinations of hashed and cached
 8548     subvols for the src and dst.
 8549     
 8550     Change-Id: I41416f9e5f2b7ea1c880d1913fdd6576da1ee868
 8551     fixes: bz#1626543
 8552     Signed-off-by: N Balachandran <nbalacha@redhat.com>
 8553 
 8554 commit bbab770e2aa527059c6590162d944a90efb459c1
 8555 Author: Atin Mukherjee <amukherj@redhat.com>
 8556 Date:   Mon Aug 19 09:40:48 2019 +0530
 8557 
 8558     tests: mark bug-1482023-snpashot-issue-with-other-processes-accessing-mounted-path.t as BRICK_MUX_BAD_TEST
 8559     
 8560     Updates: bz#1743069
 8561     Change-Id: I1eea0186ca0c1b1226f4b3d0d7c0e41fc7821cbd
 8562     Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
 8563 
 8564 commit c370c70f77079339e2cfb7f284f3a2fb13fd2f97
 8565 Author: Mohit Agrawal <moagrawal@redhat.com>
 8566 Date:   Tue Aug 13 18:45:43 2019 +0530
 8567 
 8568     rpc: glusterd start is failed and throwing an error Address already in use
 8569     
 8570     Problem: Some of the .t are failed due to bind is throwing
 8571              an error EADDRINUSE
 8572     
 8573     Solution: After killing all gluster processes .t is trying
 8574               to start glusterd but somehow if kernel has not cleaned
 8575               up resources(socket) then glusterd startup is failed due to
 8576               bind system call failure.To avoid the issue retries to call
 8577               bind 10 times to execute system call succesfully
 8578     
 8579     Change-Id: Ia5fd6b788f7b211c1508c1b7304fc08a32266629
 8580     Fixes: bz#1743020
 8581     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
 8582 
 8583 commit 29cfe643cbcf7826375a3d52ff1865706a35bc38
 8584 Author: Amar Tumballi <amarts@redhat.com>
 8585 Date:   Fri Jun 28 22:51:37 2019 +0530
 8586 
 8587     libglusterfs: remove dependency of rpc
 8588     
 8589     Goal: 'libglusterfs' files shouldn't have any dependency outside of
 8590           the tree, specially the header files, shouldn't have '#include'
 8591           from outside the tree.
 8592     
 8593     Fixes:
 8594     * Had to introduce libglusterd so, methods and structures required
 8595       for only mgmt/glusterd, and cli/ are separated from 'libglusterfs/'
 8596     * Remove rpc/xdr/gen from build, which was used mainly so
 8597       dependency for libglusterfs could be properly satisfied.
 8598     * Move rpcsvc_auth_data to client_t.h, so all dependencies could
 8599       be handled.
 8600     
 8601     Updates: bz#1636297
 8602     Change-Id: I0e80243a5a3f4615e6fac6e1b947ad08a9363fce
 8603     Signed-off-by: Amar Tumballi <amarts@redhat.com>
 8604 
 8605 commit a24288c5c9da9137eb1d5d487bc7cd11af01b88b
 8606 Author: Philip Spencer <pspencer@fields.utoronto.ca>
 8607 Date:   Tue May 7 10:19:34 2019 -0400
 8608 
 8609     mount.glusterfs: make fcache-keep-open option take a value
 8610     
 8611     Fixes: bz#1158130
 8612     Change-Id: Ifdeaed7c9fbe85f7ce421f7c89cbe7265e45f77c
 8613     Signed-off-by: Amar Tumballi <amarts@redhat.com>
 8614 
 8615 commit 8e9c53ebf16705b9a1db2fc486dc24a5cb244ddd
 8616 Author: Ravishankar N <ravishankar@redhat.com>
 8617 Date:   Tue Jul 30 17:05:22 2019 +0530
 8618 
 8619     afr: restore timestamp of parent dir during entry-heal
 8620     
 8621     Fixes: bz#1734370
 8622     Change-Id: I29e338bac62104233a6f80212df8d0fb016affda
 8623     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
 8624 
 8625 commit 23673fde22483d1b53d17ca9b7d946330eea8af3
 8626 Author: Yaniv Kaul <ykaul@redhat.com>
 8627 Date:   Tue Aug 6 23:20:02 2019 +0300
 8628 
 8629     client-handshake.c: minor changes and removal of dead code.
 8630     
 8631     - Removal of quite a bit of dead code.
 8632     - Use dict_set_str_sizen and friends where applicable.
 8633     - Moved some functions to be static and initialize values right away.
 8634     
 8635     Change-Id: Ic25b5da4028198694a0e24796dea375661eb66b9
 8636     updates: bz#1193929
 8637     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 8638 
 8639 commit d79de9976ea4332aa8f7bba4e27b0f578bb81cf0
 8640 Author: Raghavendra Talur <rtalur@redhat.com>
 8641 Date:   Thu Aug 1 16:20:33 2019 -0400
 8642 
 8643     posix: don't expect timer wheel to be inited
 8644     
 8645     Adding a timer to timer wheel should be done only after getting the
 8646     timer wheel from the ctx using the function glusterfs_ctx_tw_get(). The
 8647     function inits the wheel if not already done.
 8648     
 8649     Change-Id: I9692f84b822a02a9dc14725b7c11d26a2a634e94
 8650     Updates: #703
 8651     Signed-off-by: Raghavendra Talur <rtalur@redhat.com>
 8652 
 8653 commit 6ef9ffea7fee43ea6f59c8c36fd183f90e9c26f8
 8654 Author: N Balachandran <nbalacha@redhat.com>
 8655 Date:   Thu Jul 25 11:51:32 2019 +0530
 8656 
 8657     glusterd: create separate logdirs for cluster.rc instances
 8658     
 8659     Create a separate logdir for each host instance created by
 8660     cluster.rc. This makes it easier to determine the files
 8661     belonging to a particular instance.
 8662     
 8663     Change-Id: Ic8321f83f98995412b7d5f095b3d3f0391767a8b
 8664     Fixes: bz#1733042
 8665     Signed-off-by: N Balachandran <nbalacha@redhat.com>
 8666 
 8667 commit 089e53c7f1c32644ce2b37830b678b4c0c7071fc
 8668 Author: N Balachandran <nbalacha@redhat.com>
 8669 Date:   Fri Aug 9 14:34:22 2019 +0530
 8670 
 8671     fuse: Set limit on invalidate queue size
 8672     
 8673     If the glusterfs fuse client process is unable to
 8674     process the invalidate requests quickly enough, the
 8675     number of such requests quickly grows large enough
 8676     to use a significant amount of memory.
 8677     We are now introducing another option to set an upper
 8678     limit on these to prevent runaway memory usage.
 8679     
 8680     Change-Id: Iddfff1ee2de1466223e6717f7abd4b28ed947788
 8681     Fixes: bz#1732717
 8682     Signed-off-by: N Balachandran <nbalacha@redhat.com>
 8683 
 8684 commit 5707f2eb1c6933e70300227a6068b6b50befbb87
 8685 Author: Ashish Pandey <aspandey@redhat.com>
 8686 Date:   Thu Aug 8 11:32:56 2019 +0530
 8687 
 8688     cluster/ec: Fix coverity issue.
 8689     
 8690     Change-Id: I727287784a15d89441865de7f438002e4a370250
 8691     fixes: bz#1738763
 8692     Signed-off-by: Ashish Pandey <aspandey@redhat.com>
 8693 
 8694 commit 82ebdab2762d347549e7864658b5e20dae8379e1
 8695 Author: Amar Tumballi <amarts@redhat.com>
 8696 Date:   Thu Jul 25 13:43:37 2019 +0530
 8697 
 8698     gfapi: provide version for statedump path
 8699     
 8700     Updates: bz#1689097
 8701     Change-Id: I2e3c2a3aef17fda67d04fa9e604ea3f8335a2b0b
 8702     Signed-off-by: Amar Tumballi <amarts@redhat.com>
 8703 
 8704 commit 51237eda7c4b3846d08c5d24d1e3fe9b7ffba1d4
 8705 Author: Krutika Dhananjay <kdhananj@redhat.com>
 8706 Date:   Wed Aug 7 12:12:43 2019 +0530
 8707 
 8708     features/shard: Send correct size when reads are sent beyond file size
 8709     
 8710     Change-Id: I0cebaaf55c09eb1fb77a274268ff564e871b743b
 8711     fixes bz#1738419
 8712     Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
 8713 
 8714 commit 3cc0113ee436b63f69e16e4e0cd9443f1b3188ec
 8715 Author: Ravishankar N <ravishankar@redhat.com>
 8716 Date:   Mon Aug 12 11:29:36 2019 +0530
 8717 
 8718     tests: fix bug-880898.t crash
 8719     
 8720     Problem:
 8721     https://build.gluster.org/job/centos7-regression/7337/consoleFull
 8722     indicates the shd crashing for this .t. On looking at the core, I see
 8723     the crash is at the time of shd init and glusterfs context is null:
 8724     
 8725     (gdb) bt
 8726     (gdb) p ctx
 8727     $2 = (glusterfs_ctx_t *) 0xf00000000
 8728     
 8729     The .t is killing all gluster processes immediately after volume start,
 8730     so it looks like a race between shd coming up and it being killed.
 8731     
 8732     Fix: Kill gluster processes only after they are up and running.
 8733     
 8734     Fixes: bz#1740017
 8735     Change-Id: I7cf589201669bd9f535e968d147015dc99e9a4b6
 8736     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
 8737 
 8738 commit 40f476f40b37cc8fcc83618094d357ee643442d8
 8739 Author: Csaba Henk <csaba@redhat.com>
 8740 Date:   Wed Apr 3 14:23:06 2019 +0200
 8741 
 8742     fuse: rate limit reading from fuse device upon receiving EPERM
 8743     
 8744     Fixes: bz#1644322
 8745     Change-Id: I53e8fa362cd8c7d04fb1c4abb606a9abb642c592
 8746     Signed-off-by: Csaba Henk <csaba@redhat.com>
 8747 
 8748 commit 8ca15e3c9c72511bbe8a548a1471f8d6a5bfd709
 8749 Author: Csaba Henk <csaba@redhat.com>
 8750 Date:   Thu Jul 25 10:49:10 2019 +0200
 8751 
 8752     build: stop suppressing "Entering/Leaving direcory..." messages
 8753     
 8754     This is useful information as it allows locating build issues quickly.
 8755     
 8756     The size of the build output increases roughly by half of it, which
 8757     is a tolerable trade-off for better build time diagnostics.
 8758     
 8759     updates bz#1193929
 8760     Signed-off-by: Csaba Henk <csaba@redhat.com>
 8761     Change-Id: Id60c03bee558528ae804f037c93457f9d9e738c0
 8762 
 8763 commit b0b57af4050f19fd9b38d48cfa35c119b0ac83f4
 8764 Author: Pranith Kumar K <pkarampu@redhat.com>
 8765 Date:   Fri Aug 2 12:05:09 2019 +0530
 8766 
 8767     cluster/ec: Update lock->good_mask on parent fop failure
 8768     
 8769     When discard/truncate performs write fop, it should do so
 8770     after updating lock->good_mask to make sure readv happens
 8771     on the correct mask
 8772     
 8773     fixes bz#1727081
 8774     Change-Id: Idfef0bbcca8860d53707094722e6ba3f81c583b7
 8775     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 8776 
 8777 commit b1acf275a0779542d0ad1362e926658c037c94d8
 8778 Author: Aravinda VK <avishwan@redhat.com>
 8779 Date:   Mon Aug 5 19:00:21 2019 +0530
 8780 
 8781     geo-rep: Fix Config Get Race
 8782     
 8783     When two threads(sync jobs) in Geo-rep worker calls `gconf.get` and
 8784     `gconf.getr`(realtime) at the sametime, `getr` resets the conf object
 8785     and other one gets None. Thread Lock is introduced to fix the issue.
 8786     
 8787     ```
 8788       File "/usr/libexec/glusterfs/python/syncdaemon/syncdutils.py",
 8789       line 368, in twrap
 8790         tf(*aargs)
 8791       File "/usr/libexec/glusterfs/python/syncdaemon/master.py", line 1987,
 8792       in syncjob
 8793         po = self.sync_engine(pb, self.log_err)
 8794       File "/usr/libexec/glusterfs/python/syncdaemon/resource.py",
 8795       line 1444, in rsync
 8796         rconf.ssh_ctl_args + \
 8797     AttributeError: 'NoneType' object has no attribute 'split'
 8798     ```
 8799     
 8800     Change-Id: I9c245e5c36338265354e158f5baa32b119eb2da5
 8801     Updates: bz#1737484
 8802     Signed-off-by: Aravinda VK <avishwan@redhat.com>
 8803 
 8804 commit 30e42d14d8d1bec1883f1b7921d01b3117b1eed8
 8805 Author: Atin Mukherjee <amukherj@redhat.com>
 8806 Date:   Mon Aug 5 21:16:35 2019 +0530
 8807 
 8808     rpc/transport: have default listen-port
 8809     
 8810     With release-6, we now can have transport.socket.listen-port parameter
 8811     configurable in glusterd.vol. However the default value wasn't defined
 8812     in the code and this breaks the backward compatibility where if one has
 8813     a modified glusterd.vol file, then post upgrade the same file will be
 8814     retained and the new changes introduced as part of the release wouldn't
 8815     be available in the glusterd.vol. So it's important that for each new
 8816     options introduced in glusterd.vol file backward compatibility is
 8817     guaranteed.
 8818     
 8819     Fixes: bz#1737676
 8820     Change-Id: I776b28bff786320cda299fe673d824024dc9803e
 8821     Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
 8822 
 8823 commit e3592d7fab0dc9d3f304d3eda34ba862f2c657a6
 8824 Author: N Balachandran <nbalacha@redhat.com>
 8825 Date:   Mon Jul 29 15:20:22 2019 +0530
 8826 
 8827     cluster/dht: Log hashes in hex
 8828     
 8829     Log layout hash ranges in hex to make it easier
 8830     to compare them to the on disk xattrs.
 8831     
 8832     Change-Id: Ib75c2508bf8e0ab7f5ae26d0443ef02b792b7307
 8833     Fixes: bz#1697293
 8834     Signed-off-by: N Balachandran <nbalacha@redhat.com>
 8835 
 8836 commit 0486821b33b2ece4058a78058f9f3e2f7149d8a7
 8837 Author: Kinglong Mee <kinglongmee@gmail.com>
 8838 Date:   Mon Aug 5 11:08:02 2019 +0800
 8839 
 8840     features/utime: always update ctime at setattr
 8841     
 8842     For the nfs EXCLUSIVE mode create may sets a later time
 8843     to mtime (at verifier), it should not set to ctime for
 8844     storage.ctime does not allowed set ctime to a earlier time.
 8845     
 8846      /* Earlier, mdata was updated only if the existing time is less
 8847       * than the time to be updated. This would fail the scenarios
 8848       * where mtime can be set to any time using the syscall. Hence
 8849       * just updating without comparison. But the ctime is not
 8850       * allowed to changed to older date.
 8851       */
 8852     
 8853     According to kernel's setattr, always set ctime at setattr,
 8854     and doesnot set ctime from mtime at storage.ctime.
 8855     
 8856     Change-Id: I5cfde6cb7f8939da9617506e3dc80bd840e0d749
 8857     fixes: bz#1737288
 8858     Signed-off-by: Kinglong Mee <kinglongmee@gmail.com>
 8859 
 8860 commit 80fa433be6e399311a6bcaa1e074bc1ae65122e2
 8861 Author: Mohammed Rafi KC <rkavunga@redhat.com>
 8862 Date:   Fri Jul 5 14:12:54 2019 +0530
 8863 
 8864     graph/shd: attach volfile even if ctx->active is NULL
 8865     
 8866     While we receive a graph attach request, if ctx->active is
 8867     not set we used to fail assuming that the initilization has
 8868     not completed yet for the process start.
 8869     
 8870     Since the management connection is established, it will receive
 8871     attach request, even when ctx->active is NULL.
 8872     
 8873     Change-Id: Ied4d1ac63e6d4ced4a9405a78e1ce39f81dfd437
 8874     fixes: bz#1727256
 8875     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
 8876 
 8877 commit 8e47ae31e4477dc50eaf7604860ff8e7139ea547
 8878 Author: Mohammed Rafi KC <rkavunga@redhat.com>
 8879 Date:   Thu Jul 4 15:43:05 2019 +0530
 8880 
 8881     glusterd/shd: Return null proc if process is not running.
 8882     
 8883     We were ruturning first proc entry even if it is not running.
 8884     This was in an assumption that the process could have just
 8885     started and not updated the pidfile.
 8886     
 8887     Now we that we have introduced the states for process state,
 8888     we can take decision based on that.
 8889     
 8890     Change-Id: Ibfc11c966b0db599a8d6a08d8b975233b2bbfb8c
 8891     Fixes: bz#1728766
 8892     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
 8893 
 8894 commit 16548f11ab5433223c6d9a7e71f8dc2a925d1e5b
 8895 Author: Mohammed Rafi KC <rkavunga@redhat.com>
 8896 Date:   Wed Jul 31 15:31:34 2019 +0530
 8897 
 8898     tests/shd: Break down shd mux tests into multiple .t file
 8899     
 8900     Test file tests/basic/shd-mux.t was taking longer than 200
 8901     seconds in some iterations. So this patch is breaking the
 8902     test case to three files
 8903     
 8904     Change-Id: I1430f58798f876edf6368d6f4b8b5a75f0114c31
 8905     Updates: bz#1708929
 8906     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
 8907 
 8908 commit 2aeb9fb17087434d87497a85077073ea3bf94869
 8909 Author: Yaniv Kaul <ykaul@redhat.com>
 8910 Date:   Tue Jul 30 10:15:42 2019 +0300
 8911 
 8912     multiple files: reduce minor work under RCU_READ_LOCK
 8913     
 8914     1. Try to unlock faster - in error paths.
 8915     2. Remove memory allocations - do them before the lock.
 8916     
 8917     Change-Id: I1e9ddd80b99de45ad0f557d62a5f28951dfd54c8
 8918     updates: bz#1193929
 8919     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 8920 
 8921 commit 2dd43d3e5b5305db3a5ecafabdd9cf8e7775ec75
 8922 Author: Amar Tumballi <amarts@redhat.com>
 8923 Date:   Thu Aug 1 19:00:56 2019 +0530
 8924 
 8925     xdr: add code so we have more xdr functions covered
 8926     
 8927     Updates: bz#1693692
 8928     Change-Id: Ia10ccca5e1fed6c4269842ebb4d507662ca0f6a6
 8929     Signed-off-by: Amar Tumballi <amarts@redhat.com>
 8930 
 8931 commit ed7a3793073670e787063c47e55010fc7c963064
 8932 Author: Raghavendra Bhat <raghavendra@redhat.com>
 8933 Date:   Thu Aug 1 13:37:07 2019 -0400
 8934 
 8935     storage/posix: set the op_errno to proper errno during gfid set
 8936     
 8937     In posix_gfid_set, the proper error is not captured in one of
 8938     the failure cases.
 8939     
 8940     Change-Id: I1c13f0691a15d6893f1037b3a5fe385a99657e00
 8941     Fixes: bz#1736482
 8942     Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
 8943 
 8944 commit c927bc07385b79a4f86c339cd0e0eed361090d9b
 8945 Author: Atin Mukherjee <amukherj@redhat.com>
 8946 Date:   Mon Jul 29 11:51:07 2019 +0530
 8947 
 8948     tests: introduce BRICK_MUX_BAD_TESTS variable
 8949     
 8950     ... to ignore test failures specific to brick mux.
 8951     
 8952     Change-Id: I3114f34f194f18e7fc8c02a117b295f086721226
 8953     updates: bz#1193929
 8954     Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
 8955 
 8956 commit d072dc09f5bc6431ba31244cf7e9eba1b6e85229
 8957 Author: Susant Palai <spalai@redhat.com>
 8958 Date:   Mon Jul 22 15:20:43 2019 +0530
 8959 
 8960     locks/fencing: Address hang while lock preemption
 8961     
 8962     The fop_wind_count can go negative when fencing is enabled
 8963     on unwind path of the IO leading to hang.
 8964     
 8965     Also changed code so that fop_wind_count needs to be maintained only
 8966     till fencing is enabled on the file.
 8967     
 8968     updates: bz#1717824
 8969     Change-Id: Icd04b42bc16cd3d50eaa581ee57233910194f480
 8970     Signed-off-by: Susant Palai <spalai@redhat.com>
 8971 
 8972 commit 84f7794547522463841068063b22fd3a8d8fca2b
 8973 Author: Kotresh HR <khiremat@redhat.com>
 8974 Date:   Wed Jul 31 15:40:55 2019 +0530
 8975 
 8976     geo-rep: Fix mount broker setup issue
 8977     
 8978     Even the use builtin 'type' command as in patch [1]
 8979     causes issues if argument in question is not part of PATH
 8980     environment variable for that user. This patch fixes the
 8981     same by doing source /etc/profile. This was already being
 8982     used in another part of script.
 8983     
 8984     [1] https://review.gluster.org/23089
 8985     
 8986     Change-Id: Iceb78835967ec6a4350983eec9af28398410c002
 8987     fixes: bz#1734738
 8988     Signed-off-by: Kotresh HR <khiremat@redhat.com>
 8989 
 8990 commit 756f903e2adeaa520c848aa3c0664573cb584d07
 8991 Author: Soumya Koduri <skoduri@redhat.com>
 8992 Date:   Thu Jul 25 12:56:12 2019 +0530
 8993 
 8994     gfapi: Fix deadlock while processing upcall
 8995     
 8996     As mentioned in bug1733166, there could be potential deadlock
 8997     while processing upcalls depending on how each xlator choose
 8998     to act on it. The right way of fixing such issues
 8999     is to change rpc callback communication process.
 9000     - https://github.com/gluster/glusterfs/issues/697
 9001     
 9002     Till then, making changes in gfapi layer to avoid any I/O
 9003     processing.
 9004     
 9005     Change-Id: I2079e95339e5d761d5060707f4555cfacab95c83
 9006     fixes: bz#1733166
 9007     Signed-off-by: Soumya Koduri <skoduri@redhat.com>
 9008 
 9009 commit 00084d5f0c84995d14fb3b626af227f60426e885
 9010 Author: Amar Tumballi <amarts@redhat.com>
 9011 Date:   Thu Aug 1 12:13:05 2019 +0530
 9012 
 9013     lcov: check for zerofill/discard fops on arbiter
 9014     
 9015     Updates: bz#1693692
 9016     Change-Id: I145df4c0d0b0ce738f1d34b02341ec606e38522e
 9017     Signed-off-by: Amar Tumballi <amarts@redhat.com>
 9018 
 9019 commit d68e426263708334945dff69b88218ed71a48033
 9020 Author: Yaniv Kaul <ykaul@redhat.com>
 9021 Date:   Sat Jun 8 01:22:27 2019 +0300
 9022 
 9023     Multiple files: get trivial stuff done before lock
 9024     
 9025     Initialize a dictionary for example seems to be prefectly fine to be done
 9026     before taking a lock.
 9027     
 9028     Change-Id: Ib29516c4efa8f0e2b526d512beab488fcd16d2e7
 9029     updates: bz#1193929
 9030     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 9031 
 9032 commit d7e449b0da35a30af876aef1bea1fb096d0c06dc
 9033 Author: Kotresh HR <khiremat@redhat.com>
 9034 Date:   Mon Jul 29 16:22:10 2019 +0530
 9035 
 9036     posix/ctime: Fix race during lookup ctime xattr heal
 9037     
 9038     Problem:
 9039     Ctime heals the ctime xattr ("trusted.glusterfs.mdata") in lookup
 9040     if it's not present. In a multi client scenario, there is a race
 9041     which results in updating the ctime xattr to older value.
 9042     
 9043     e.g. Let c1 and c2 be two clients and file1 be the file which
 9044     doesn't have the ctime xattr. Let the ctime of file1 be t1.
 9045     (from backend, ctime heals time attributes from backend when not present).
 9046     
 9047     Now following operations are done on mount
 9048     c1 -> ls -l /mnt/file1  |   c2 -> ls -l /mnt/file1;echo "append" >> /mnt/file1;
 9049     
 9050     The race is that the both c1 and c2 didn't fetch the ctime xattr in lookup,
 9051     so both of them tries to heal ctime to time 't1'. If c2 wins the race and
 9052     appends the file before c1 heals it, it sets the time to 't1' and updates
 9053     it to 't2' (because of append). Now c1 proceeds to heal and sets it to 't1'
 9054     which is incorrect.
 9055     
 9056     Solution:
 9057     Compare the times during heal and only update the larger time. This is the
 9058     general approach used in ctime feature but got missed with healing legacy
 9059     files.
 9060     
 9061     fixes: bz#1734299
 9062     Change-Id: I930bda192c64c3d49d0aed431ce23d3bc57e51b7
 9063     Signed-off-by: Kotresh HR <khiremat@redhat.com>
 9064 
 9065 commit 7124fc6d01aa23e9a218d0ce608242d0ae320193
 9066 Author: Amar Tumballi <amarts@redhat.com>
 9067 Date:   Thu Jul 25 11:00:29 2019 +0530
 9068 
 9069     gfapi: increase function-coverage
 9070     
 9071     * Add few more mgmt functions to the coverage
 9072     * While testing mgmt function, found an issue, where if the
 9073       'glfs_set_volfile_server()' is not called before calling
 9074       'glfs_unset_volfile_server()', unset would cause a crash.
 9075       Null check of few variables fixes the issue, which is handled
 9076       in this patch itself.
 9077     * Added a test for volfile API
 9078     
 9079     Updates: bz#1693692
 9080     Change-Id: Iba151f8da1b64107e2f436ddbfef9da45b1c1588
 9081     Signed-off-by: Amar Tumballi <amarts@redhat.com>
 9082 
 9083 commit a31fad885c30cbc1bea652349c7d52bac1414c08
 9084 Author: Amar Tumballi <amarts@redhat.com>
 9085 Date:   Tue Jul 30 17:12:58 2019 +0530
 9086 
 9087     tests: heal-info add --xml option for more coverage
 9088     
 9089     Updates: bz#1693692
 9090     Change-Id: I13de7cf4c380b9663e6258e2fd62ce1f180591b4
 9091     Signed-off-by: Amar Tumballi <amarts@redhat.com>
 9092 
 9093 commit 1b19ee85d5eec97b21fbd3fe6d283ec38a23028b
 9094 Author: Amar Tumballi <amarts@redhat.com>
 9095 Date:   Thu Jul 25 10:55:01 2019 +0530
 9096 
 9097     error-gen: increase coverage by reducing error-rate
 9098     
 9099     Updates: bz#1693692
 9100     Change-Id: I07371ea1e2613fed6dd9ea67c40cbb3ebcb9387e
 9101     Signed-off-by: Amar Tumballi <amarts@redhat.com>
 9102 
 9103 commit 363527ebc4140e99f182770915b72acdcb5c2bdf
 9104 Author: Amar Tumballi <amarts@redhat.com>
 9105 Date:   Mon Jul 29 15:07:58 2019 +0530
 9106 
 9107     trace: add more coverage by testing it with glfs-coverage too.
 9108     
 9109     make sure to provide 'log-file' option, so we can see the logs.
 9110     This test does test volgen inserting the trace xlator in server
 9111     graph.
 9112     
 9113     Updates: bz#1693692
 9114     Change-Id: I26c736b04376674b4c094d48060660421e6c983c
 9115     Signed-off-by: Amar Tumballi <amarts@redhat.com>
 9116 
 9117 commit 22bd0570c04cb94d1a954b56d11db61790ff20b2
 9118 Author: Ashish Pandey <aspandey@redhat.com>
 9119 Date:   Tue Jul 30 10:32:39 2019 +0530
 9120 
 9121     cluster/ec: Create heal task with heal process id
 9122     
 9123     Problem:
 9124     ec_data_undo_pending calls syncop_fxattrop->SYNCOP without
 9125     a frame. In this case SYNCOP gets the frame of the task.
 9126     However, when we create a synctask for heal we provide
 9127     frame as NULL.
 9128     Now, if the read-only feature is ON, it will receive the
 9129     process ID of the shd as 0 and will consider that it as
 9130     not an internal process. This will prevent healing of a
 9131     file with "Read-only file system" error message log.
 9132     
 9133     Solution:
 9134     While launching heal, create a synctask using frame and set
 9135     process id of the SHD which is -6.
 9136     
 9137     Change-Id: I37195399c85de322cbcac75633888922c4e3db4a
 9138     Fixes: bz#1734252
 9139 
 9140 commit 065b10beaae8ea7390c2e3059044313aef8a5fea
 9141 Author: Pranith Kumar K <pkarampu@redhat.com>
 9142 Date:   Mon Jul 29 14:08:37 2019 +0530
 9143 
 9144     cluster/ec: Fix reopen flags to avoid misbehavior
 9145     
 9146     Problem:
 9147     when a file needs to be re-opened O_APPEND and O_EXCL
 9148     flags are not filtered in EC.
 9149     
 9150     - O_APPEND should be filtered because EC doesn't send O_APPEND below EC for
 9151     open to make sure writes happen on the individual fragments instead of at the
 9152     end of the file.
 9153     
 9154     - O_EXCL should be filtered because shd could have created the file so even
 9155     when file exists open should succeed
 9156     
 9157     - O_CREAT should be filtered because open happens with gfid as parameter. So
 9158     open fop will create just the gfid which will lead to problems.
 9159     
 9160     Fix:
 9161     Filter out these two flags in reopen.
 9162     
 9163     Change-Id: Ia280470fcb5188a09caa07bf665a2a94bce23bc4
 9164     Fixes: bz#1733935
 9165     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 9166 
 9167 commit d7001336c38bb9516645bbc0fd07a8c7f1f8b3f8
 9168 Author: Yaniv Kaul <ykaul@redhat.com>
 9169 Date:   Wed Jul 3 10:07:29 2019 +0300
 9170 
 9171     dict.c: dict_unserialize() can use dict_addn()
 9172     
 9173     Since we know the keylen, we can use dict_addn() instead of dict_add().
 9174     
 9175     Change-Id: Iebe77faa1a20ec132cba5fb85b3d2ea63339181d
 9176     updates: bz#1193929
 9177     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 9178 
 9179 commit 799edc73c3d4f694c365c6a7c27c9ab8eed5f260
 9180 Author: Xiubo Li <xiubli@redhat.com>
 9181 Date:   Fri Jul 26 12:34:52 2019 +0800
 9182 
 9183     event: rename event_XXX with gf_ prefixed
 9184     
 9185     I hit one crash issue when using the libgfapi.
 9186     
 9187     In the libgfapi it will call glfs_poller() --> event_dispatch()
 9188     in file api/src/glfs.c:721, and the event_dispatch() is defined
 9189     by libgluster locally, the problem is the name of event_dispatch()
 9190     is the extremly the same with the one from libevent package form
 9191     the OS.
 9192     
 9193     For example, if a executable program Foo, which will also use and
 9194     link the libevent and the libgfapi at the same time, I can hit the
 9195     crash, like:
 9196     
 9197     kernel: glfs_glfspoll[68486]: segfault at 1c0 ip 00007fef006fd2b8 sp
 9198     00007feeeaffce30 error 4 in libevent-2.0.so.5.1.9[7fef006ed000+46000]
 9199     
 9200     The link for Foo is:
 9201     lib_foo_LADD = -levent $(GFAPI_LIBS)
 9202     It will crash.
 9203     
 9204     This is because the glfs_poller() is calling the event_dispatch() from
 9205     the libevent, not the libglsuter.
 9206     
 9207     The gfapi link info :
 9208     GFAPI_LIBS = -lacl -lgfapi -lglusterfs -lgfrpc -lgfxdr -luuid
 9209     
 9210     If I link Foo like:
 9211     lib_foo_LADD = $(GFAPI_LIBS) -levent
 9212     It will works well without any problem.
 9213     
 9214     And if Foo call one private lib, such as handler_glfs.so, and the
 9215     handler_glfs.so will link the GFAPI_LIBS directly, while the Foo won't
 9216     and it will dlopen(handler_glfs.so), then the crash will be hit everytime.
 9217     
 9218     The link info will be:
 9219     foo_LADD = -levent
 9220     libhandler_glfs_LIBADD = $(GFAPI_LIBS)
 9221     
 9222     I can avoid the crash temporarily by linking the GFAPI_LIBS in Foo too like:
 9223     foo_LADD = $(GFAPI_LIBS) -levent
 9224     libhandler_glfs_LIBADD = $(GFAPI_LIBS)
 9225     
 9226     But this is ugly since the Foo won't use any APIs from the GFAPI_LIBS.
 9227     
 9228     And in some cases when the --as-needed link option is added(on many dists
 9229     it is added as default), then the crash is back again, the above workaround
 9230     won't work.
 9231     
 9232     Fixes: #699
 9233     
 9234     Change-Id: I38f0200b941bd1cff4bf3066fca2fc1f9a5263aa
 9235     Signed-off-by: Xiubo Li <xiubli@redhat.com>
 9236 
 9237 commit 26b9c85861f058a4a8335b28f49410ba061e7c1a
 9238 Author: Yaniv Kaul <ykaul@redhat.com>
 9239 Date:   Thu Jul 25 23:36:56 2019 +0300
 9240 
 9241     glusterd: write voldir once in glusterd-store and don't attempt again.
 9242     
 9243     glusterd_store_brickinfos() is calling per each brick the
 9244     function glusterd_store_brickinfo().
 9245     In it, we call:
 9246     
 9247      ret = glusterd_store_create_brick_dir(volinfo);
 9248     
 9249     However, volinfo is the same for all those bricks - no need to again
 9250     and again call it (which tries to mkdir that dir).
 9251     We can do it once above the loops in glusterd_store_brickinfos()
 9252     
 9253     While at, combine two similar functions that write additional dirs.
 9254     
 9255     Change-Id: I5858cf7783f088ea13a8fa20115118efa816f4cb
 9256     updates: bz#1193929
 9257     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 9258 
 9259 commit 11d52142f67638209be43007d993dc137412aa42
 9260 Author: Pranith Kumar K <pkarampu@redhat.com>
 9261 Date:   Thu Jul 18 11:25:31 2019 +0530
 9262 
 9263     cluster/ec: Always read from good-mask
 9264     
 9265     There are cases where fop->mask may have fop->healing added
 9266     and readv shouldn't be wound on fop->healing. To avoid this
 9267     always wind readv to lock->good_mask
 9268     
 9269     fixes bz#1727081
 9270     Change-Id: I2226ef0229daf5ff315d51e868b980ee48060b87
 9271     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 9272 
 9273 commit 2af877cec513b1fca5a9f27b338117bacd5ab2f1
 9274 Author: Csaba Henk <csaba@redhat.com>
 9275 Date:   Tue Jul 9 11:06:49 2019 +0200
 9276 
 9277     fuse: add missing GF_FREE to fuse_interrupt
 9278     
 9279     Change-Id: Id7e003e4a53d0a0057c1c84e1cd704c80a6cb015
 9280     Fixes: bz#1728047
 9281     Signed-off-by: Csaba Henk <csaba@redhat.com>
 9282 
 9283 commit f024178b7cd4f884e23045c77dc4e09b89b98bb7
 9284 Author: Amar Tumballi <amarts@redhat.com>
 9285 Date:   Thu Jul 25 12:29:15 2019 +0530
 9286 
 9287     quiesce: add missing fops
 9288     
 9289     Updates: bz#1693692
 9290     Change-Id: I4f005e7168c201709a85db443d643b81e6d3d282
 9291     Signed-off-by: Amar Tumballi <amarts@redhat.com>
 9292 
 9293 commit 84a55090123a7e3124100e5564da8c521c3c22ab
 9294 Author: Rinku Kothiya <rkothiya@redhat.com>
 9295 Date:   Tue Jul 23 13:16:04 2019 +0000
 9296 
 9297     [core] fix return of local in  __nlc_inode_ctx_get
 9298     
 9299     __nlc_inode_ctx_get assigns a value to nlc_pe_p which is never used by
 9300     its parent function or any of the predecessor hence remove the
 9301     assignment and also that function argument as it is not being used
 9302     anywhere.
 9303     
 9304     fixes: bz#1732496
 9305     
 9306     Change-Id: I5b950e1e251bd50a646616da872a4efe9d2ff8c9
 9307     Signed-off-by: Rinku Kothiya <rkothiya@redhat.com>
 9308 
 9309 commit b01a43586c5abc23a874e5528a063c508f952cbd
 9310 Author: Xavi Hernandez <xhernandez@redhat.com>
 9311 Date:   Wed Jul 17 14:50:22 2019 +0200
 9312 
 9313     cluster/ec: fix EIO error for concurrent writes on sparse files
 9314     
 9315     EC doesn't allow concurrent writes on overlapping areas, they are
 9316     serialized. However non-overlapping writes are serviced in parallel.
 9317     When a write is not aligned, EC first needs to read the entire chunk
 9318     from disk, apply the modified fragment and write it again.
 9319     
 9320     The problem appears on sparse files because a write to an offset
 9321     implicitly creates data on offsets below it (so, in some way, they
 9322     are overlapping). For example, if a file is empty and we read 10 bytes
 9323     from offset 10, read() will return 0 bytes. Now, if we write one byte
 9324     at offset 1M and retry the same read, the system call will return 10
 9325     bytes (all containing 0's).
 9326     
 9327     So if we have two writes, the first one at offset 10 and the second one
 9328     at offset 1M, EC will send both in parallel because they do not overlap.
 9329     However, the first one will try to read missing data from the first chunk
 9330     (i.e. offsets 0 to 9) to recombine the entire chunk and do the final write.
 9331     This read will happen in parallel with the write to 1M. What could happen
 9332     is that half of the bricks process the write before the read, and the
 9333     half do the read before the write. Some bricks will return 10 bytes of
 9334     data while the otherw will return 0 bytes (because the file on the brick
 9335     has not been expanded yet).
 9336     
 9337     When EC tries to recombine the answers from the bricks, it can't, because
 9338     it needs more than half consistent answers to recover the data. So this
 9339     read fails with EIO error. This error is propagated to the parent write,
 9340     which is aborted and EIO is returned to the application.
 9341     
 9342     The issue happened because EC assumed that a write to a given offset
 9343     implies that offsets below it exist.
 9344     
 9345     This fix prevents the read of the chunk from bricks if the current size
 9346     of the file is smaller than the read chunk offset. This size is
 9347     correctly tracked, so this fixes the issue.
 9348     
 9349     Also modifying ec-stripe.t file for Test #13 within it.
 9350     In this patch, if a file size is less than the offset we are writing, we
 9351     fill zeros in head and tail and do not consider it strip cache miss.
 9352     That actually make sense as we know what data that part holds and there is
 9353     no need of reading it from bricks.
 9354     
 9355     Change-Id: Ic342e8c35c555b8534109e9314c9a0710b6225d6
 9356     Fixes: bz#1730715
 9357     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
 9358 
 9359 commit dafab9fd5e4afe0f1cdd8d2e3ce7b4037faf4830
 9360 Author: MPLX <mplx+redhat@donotreply.at>
 9361 Date:   Thu Jul 18 12:13:09 2019 +0530
 9362 
 9363     extras/mount-shared-storage.sh: consider options to mount
 9364     
 9365     If different mount options are given to glusterfs in /etc/fstab, it
 9366     was not considered in the script.
 9367     
 9368     Fixes: bz#1690454
 9369     Change-Id: If829b3d6ab79755647626edcb31f38a3716989db
 9370     Signed-off-by: Amar Tumballi <amarts@redhat.com>
 9371 
 9372 commit 8a84a2ca7e70c0682d3f0f1f83892914725db59c
 9373 Author: Prasanna Kumar Kalever <prasanna.kalever@redhat.com>
 9374 Date:   Thu Jun 27 13:18:32 2019 +0530
 9375 
 9376     gluster-block: tuning perf options
 9377     
 9378     As per the perf experiment run by Elvir, with NVME devices used for OCP (gluster)
 9379     it was seen that particularly read operations (read/randread) benefited from
 9380     these options.
 9381     
 9382     Change-Id: Ibec4b96afd28e6f7e757b6ef203ccdbc0d9854d5
 9383     Fixes: bz#1727852
 9384     Signed-off-by: Prasanna Kumar Kalever <prasanna.kalever@redhat.com>
 9385 
 9386 commit 70a735b75ac8ca942b212cc1c56441e9387f30e6
 9387 Author: Sanju Rakonde <srakonde@redhat.com>
 9388 Date:   Mon Jul 8 11:34:49 2019 +0530
 9389 
 9390     core: use more restrictive mode while creating the directories
 9391     
 9392     fixes: bz#1724024
 9393     
 9394     Change-Id: I539fb7248b2cfc037ec29f1413ea648f9ec21ef2
 9395     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
 9396 
 9397 commit 2aa731a259ea457c07494e3c3edf6d5f7c02fe77
 9398 Author: Kotresh HR <khiremat@redhat.com>
 9399 Date:   Mon Jul 22 17:35:21 2019 +0530
 9400 
 9401     geo-rep: Fix mount broker setup issue
 9402     
 9403     The patch [1] added validation in gverify.sh to check if the gluster
 9404     binary on slave by executing gluster directly on slave.  But for
 9405     non-root users, even though gluster binary is present, path is not
 9406     found when executed via ssh. Hence validate the gluster binary using
 9407     bash builtin 'type' command.
 9408     
 9409     [1] https://review.gluster.org/19224
 9410     
 9411     Change-Id: I93ca62c0c5b1e16263e586ddbbca8407d60ca126
 9412     fixes: bz#1731920
 9413     Signed-off-by: Kotresh HR <khiremat@redhat.com>
 9414 
 9415 commit f4480d6369098869a5a5f53a39ea0da7f152d227
 9416 Author: Amar Tumballi <amarts@redhat.com>
 9417 Date:   Mon Jul 22 10:23:56 2019 +0530
 9418 
 9419     backup-volfile-server: ignore duplicate entries
 9420     
 9421     Erroring out for duplicate entries for backup_volfile_servers would
 9422     cause failures for some script. This option is harmless if it is
 9423     duplicate, we just need to ignore this.
 9424     
 9425     Fixes: bz#1730953
 9426     Change-Id: I950064501a7bb61b90e99f2c079564f9c923d4c1
 9427     Signed-off-by: Amar Tumballi <amarts@redhat.com>
 9428 
 9429 commit 8f7112264f76306a12eb29fffc458ef4f657689e
 9430 Author: Pranith Kumar K <pkarampu@redhat.com>
 9431 Date:   Mon Jul 22 20:55:33 2019 +0530
 9432 
 9433     features/utime: Fix mem_put crash
 9434     
 9435     Problem:
 9436     When frame->local is not null FRAME_DESTROY calls mem_put on it.
 9437     Since the stub is already destroyed in call_resume(), it leads
 9438     to crash
 9439     
 9440     Fix:
 9441     Set frame->local to NULL before calling call_resume()
 9442     
 9443     fixes: bz#1593542
 9444     Change-Id: I0f8adf406f4cefdb89d7624ba7a9d9c2eedfb1de
 9445     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
 9446 
 9447 commit 5c20a688fd2408ff8dcc66fd8ebc935e228dd803
 9448 Author: Yaniv Kaul <ykaul@redhat.com>
 9449 Date:   Sun Jun 2 21:14:18 2019 +0300
 9450 
 9451     (multiple files) use dict_allocate_and_serialize() where applicable.
 9452     
 9453     This function does length, allocation and serialization for you.
 9454     
 9455     Change-Id: I142a259952a2fe83dd719442afaefe4a43a8e55e
 9456     updates: bz#1193929
 9457     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 9458 
 9459 commit 06e92a2ee437c1a81c815129b1d188af0b4fa84e
 9460 Author: Kotresh HR <khiremat@redhat.com>
 9461 Date:   Mon Jun 24 13:06:49 2019 +0530
 9462 
 9463     ctime: Set mdata xattr on legacy files
 9464     
 9465     Problem:
 9466     The files which were created before ctime enabled would not
 9467     have "trusted.glusterfs.mdata"(stores time attributes) xattr.
 9468     Upon fops which modifies either ctime or mtime, the xattr
 9469     gets created with latest ctime, mtime and atime, which is
 9470     incorrect. It should update only the corresponding time
 9471     attribute and rest from backend
 9472     
 9473     Solution:
 9474     Creating xattr with values from brick is not possible as
 9475     each brick of replica set would have different times.
 9476     So create the xattr upon successful lookup if the xattr
 9477     is not created
 9478     
 9479     Note To Reviewers:
 9480     The time attributes used to set xattr is got from successful
 9481     lookup. Instead of sending the whole iatt over the wire via
 9482     setxattr, a structure called mdata_iatt is sent. The mdata_iatt
 9483     contains only time attributes.
 9484     
 9485     Change-Id: I5e535631ddef04195361ae0364336410a2895dd4
 9486     fixes: bz#1593542
 9487     Signed-off-by: Kotresh HR <khiremat@redhat.com>
 9488 
 9489 commit 74f124619a71df9bdc5ae9fbc07bc19db05bc1d2
 9490 Author: Shwetha K Acharya <sacharya@redhat.com>
 9491 Date:   Tue Jul 2 15:00:25 2019 +0530
 9492 
 9493     geo-rep: Test case for upgrading config file
 9494     
 9495     Added test case for the patch
 9496     https://review.gluster.org/#/c/glusterfs/+/22894/4
 9497     
 9498     Also updated if else structure in gsyncdconfig.py to avoid
 9499     repeated occurance of values in new configfile.
 9500     
 9501     fixes: bz#1707731
 9502     Change-Id: If97e1d37ac52dbd17d47be6cb659fc5a3ccab6d7
 9503     Signed-off-by: Shwetha K Acharya <sacharya@redhat.com>
 9504 
 9505 commit 41f72ac9453c24bea85709203948c41c5772edec
 9506 Author: Susant Palai <spalai@redhat.com>
 9507 Date:   Tue Jul 16 10:31:46 2019 +0530
 9508 
 9509     dht: log getxattr failure for node-uuid at "DEBUG"
 9510     
 9511     There are two ways to fetch node-uuid information from dht.
 9512     
 9513     1 - #define GF_XATTR_LIST_NODE_UUIDS_KEY "trusted.glusterfs.list-node-uuids"
 9514     This key is used by AFR.
 9515     
 9516     2 - #define GF_REBAL_FIND_LOCAL_SUBVOL "glusterfs.find-local-subvol"
 9517     This key is used for non-afr volume type.
 9518     
 9519     We do two getxattr operations. First on the #1 key followed by on #2 if
 9520     getxattr on #1 key fails.
 9521     
 9522     Since the parent function "dht_init_local_subvols_and_nodeuuids" logs failure,
 9523     moving the log-level to DEBUG in dht_find_local_subvol_cbk.
 9524     
 9525     fixes: bz#1730175
 9526     Change-Id: I4d88244dc26587b111ca5b00d4c00118efdaac14
 9527     Signed-off-by: Susant Palai <spalai@redhat.com>
 9528 
 9529 commit 477bbe36dc2a05e00ffb137450fa18a0aba8bf6a
 9530 Author: Kinglong Mee <kinglongmee@gmail.com>
 9531 Date:   Thu Jul 11 18:57:13 2019 +0800
 9532 
 9533     cluster/ec: skip updating ctx->loc again when ec_fix_open/opendir
 9534     
 9535     The ec_manager_open/opendir memsets ctx->loc which causes
 9536     memory/inode leak, and ec_fheal uses ctx->loc out of fd->lock
 9537     that loc_copy may copy bad data when memset it.
 9538     
 9539     This patch skips updating ctx->loc when it is initilizaed.
 9540     With it, ctx->loc is filled once, and never updated.
 9541     
 9542     Change-Id: I3bf5ffce4caf4c1c667f7acaa14b451d37a3550a
 9543     fixes: bz#1729772
 9544     Signed-off-by: Kinglong Mee <mijinlong@horiscale.com>
 9545 
 9546 commit 11de436ba7ef4cc1f583609f2482b1008607d410
 9547 Author: Yaniv Kaul <ykaul@redhat.com>
 9548 Date:   Wed Jul 3 20:47:52 2019 +0300
 9549 
 9550     logging.c: check for skip_logging() in _gf_slog(), _gf_smsg()
 9551     
 9552     Before processing anything, check if we actually need to do any logging,
 9553     based on the log level.
 9554     
 9555     Change-Id: Iec5b9869831e1bdf22d12becc5e3d3ece87906a7
 9556     updates: bz#1193929
 9557     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 9558 
 9559 commit 47e9658fd9014d9a4b67718aa9fbbdf11bf93c0e
 9560 Author: Kinglong Mee <kinglongmee@gmail.com>
 9561 Date:   Mon Jul 8 21:13:28 2019 +0800
 9562 
 9563     cluster/ec: inherit healing from lock when it has info
 9564     
 9565     If lock has info, fop should inherit healing mask from it.
 9566     Otherwise, fop cannot inherit right healing when changed_flags is zero.
 9567     
 9568     Change-Id: Ife80c9169d2c555024347a20300b0583f7e8a87f
 9569     fixes: bz#1727081
 9570     Signed-off-by: Kinglong Mee <mijinlong@horiscale.com>
 9571 
 9572 commit 7162c8ad9d404773656b4eb02c8fc17f067e7699
 9573 Author: Yaniv Kaul <ykaul@redhat.com>
 9574 Date:   Wed Jun 5 20:16:40 2019 +0300
 9575 
 9576     cli: defer create_frame() (and dict creation) to later stages.
 9577     
 9578     Where possible, defer create_frame() - whenever possible, after
 9579     command line verification, for example.
 9580     
 9581     Change-Id: Id6606e90e7ea6190f30b225c4733b229c519bb2f
 9582     updates: bz#1193929
 9583     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 9584 
 9585 commit 6bf9637a93011298d032332ca93009ba4e377e46
 9586 Author: Homma <homma@allworks.co.jp>
 9587 Date:   Fri Jul 5 16:10:41 2019 +0530
 9588 
 9589     system/posix-acl: update ctx only if iatt is non-NULL
 9590     
 9591     We need to safe-guard against possible zero'ing out of iatt
 9592     structure in acl ctx, which can cause many issues.
 9593     
 9594     fixes: bz#1668286
 9595     Change-Id: Ie81a57d7453a6624078de3be8c0845bf4d432773
 9596     Signed-off-by: Amar Tumballi <amarts@redhat.com>
 9597 
 9598 commit 30f54b702a82c45a31aeb58fabe5b79fb62cc37b
 9599 Author: karthik-us <ksubrahm@redhat.com>
 9600 Date:   Mon Jul 15 11:51:58 2019 +0530
 9601 
 9602     tests: Fix bug-1717819-metadata-split-brain-detection.t failure
 9603     
 9604     Problem:
 9605     tests/bugs/replicate/bug-1717819-metadata-split-brain-detection.t fails
 9606     intermittently in test cases #49 & #50, which compare the values of the
 9607     user set xattr values after enabling the heal. We are not waiting for
 9608     the heal to complete before comparing those values, which might lead
 9609     those tests to fail.
 9610     
 9611     Fix:
 9612     Wait till the HEAL-TIMEOUT before comparing the xattr values.
 9613     Also cheking for the shd to come up and the bricks to connect to the shd
 9614     process in another case.
 9615     
 9616     Change-Id: I0e245b328da9df23ce70c5300278fad1c1d9f7ff
 9617     Fixes: bz#1729847
 9618     Signed-off-by: karthik-us <ksubrahm@redhat.com>
 9619 
 9620 commit ea40178dcc540535e1059a9433f74a30de76f03b
 9621 Author: Yaniv Kaul <ykaul@redhat.com>
 9622 Date:   Sat Apr 20 19:11:33 2019 +0300
 9623 
 9624     dht-common.h: reorder variables to reduce padding.
 9625     
 9626     Manually added '-Wpadded' to get warnings on padding, and reordered
 9627     structs to reduce most of them.
 9628     
 9629     Change-Id: I0c505fcb3dfef76399ac9d5d33bfb235354532de
 9630     updates: bz#1193929
 9631     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 9632 
 9633 commit a263b22904036c04b52b1555239ae841f676a564
 9634 Author: Xavi Hernandez <xhernandez@redhat.com>
 9635 Date:   Thu Jul 4 13:21:33 2019 +0200
 9636 
 9637     core: fix deadlock between statedump and fd_anonymous()
 9638     
 9639     There exists a deadlock between statedump generation and fd_anonymous()
 9640     function because they are acquiring inode table lock and inode lock in
 9641     reverse order.
 9642     
 9643     This patch modifies fd_anonymous() so that it takes inode lock only when
 9644     it's really necessary, avoiding the deadlock.
 9645     
 9646     Change-Id: I24355447f0ea1b39e2546782ad07f0512cc381e7
 9647     Fixes: bz#1727068
 9648     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
 9649 
 9650 commit d509f480b2d85dc86f027edc4c9badf79b5eb62f
 9651 Author: Liguang Li <liguang.lee6@gmail.com>
 9652 Date:   Fri Jun 21 12:18:58 2019 +0800
 9653 
 9654     Detach iot_worker to release its resources
 9655     
 9656     When iot_worker terminates, its resources have not been reaped, which
 9657     will consumes lots of memory.
 9658     
 9659     Detach iot_worker to automically release its resources back to the
 9660     system.
 9661     
 9662     fixes: bz#1729107
 9663     Change-Id: I71fabb2940e76ad54dc56b4c41aeeead2644b8bb
 9664     Signed-off-by: Liguang Li <liguang.lee6@gmail.com>
 9665 
 9666 commit b917974ee922d7a2e079692ad7d6f61f900b37b2
 9667 Author: Sanju Rakonde <srakonde@redhat.com>
 9668 Date:   Fri Jul 12 16:28:04 2019 +0530
 9669 
 9670     glusterd: do not mark skip_locking as true for geo-rep operations
 9671     
 9672     We need to send the commit req to peers in case of geo-rep
 9673     operations even though it is a no volname operation. In commit
 9674     phase peers try to set the txn_opinfo which will fail because
 9675     it is a no volname operation where we don't require a commit
 9676     phase. We mark skip_locking as true for no volname operations,
 9677     but we have to give an exception to geo-rep operations, so that
 9678     they can set txn_opinfo in commit phase.
 9679     
 9680     Please refer to detailed RCA at the bug: 1729463
 9681     
 9682     fixes: bz#1729463
 9683     
 9684     Change-Id: I9f2478b12a281f6e052035c0563c40543493a3fc
 9685     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
 9686 
 9687 commit 6d94a332ebafbda598d50efcd64429af563dae30
 9688 Author: Aravinda VK <avishwan@redhat.com>
 9689 Date:   Wed Jul 10 12:34:34 2019 +0530
 9690 
 9691     Fix spelling errors
 9692     
 9693     Fixes: bz#1728554
 9694     Change-Id: I88357aed7c14988a12616035c3738c32c09a8f9a
 9695     Signed-off-by: Patrick Matthäi <pmatthaei@debian.org>
 9696     Signed-off-by: Aravinda VK <avishwan@redhat.com>
 9697 
 9698 commit 8f37561d665b6a70b2b4bd8afab12affb1805148
 9699 Author: Amar Tumballi <amarts@redhat.com>
 9700 Date:   Fri Jul 12 16:12:18 2019 +0530
 9701 
 9702     ibverbs/rdma: remove from build
 9703     
 9704     We have proposed about this an year ago, and with recent smoke
 9705     failures, it looks like the right time to take such call.
 9706     
 9707     ref: https://lists.gluster.org/pipermail/gluster-users/2018-July/034400.html
 9708     
 9709     With this, glusterfs-8.0 wouldn't have rdma feature, and would
 9710     allow some modularity changes possible with rpc layer (as we would
 9711     have just 1 transport)
 9712     
 9713     Updates: bz#1635688
 9714     Change-Id: Ia277dca4d4b1f0cffae20819024a52b075b775e5
 9715     Signed-off-by: Amar Tumballi <amarts@redhat.com>
 9716 
 9717 commit 8e795617fd6f5193d0d52a336059ce1a28108c0e
 9718 Author: Raghavendra Bhat <raghavendra@redhat.com>
 9719 Date:   Tue Jul 2 16:50:23 2019 -0400
 9720 
 9721     features/snapview-server: obtain the list of snapshots inside the lock
 9722     
 9723     The current list of snapshots from priv->dirents is obtained outside
 9724     the lock.
 9725     
 9726     Change-Id: I8876ec0a38308da5db058397382fbc82cc7ac177
 9727     Fixes: bz#1726783
 9728 
 9729 commit 3ef02e82f84e20edaa8ad4cbe921108319ec8ba2
 9730 Author: karthik-us <ksubrahm@redhat.com>
 9731 Date:   Wed Jun 26 12:19:07 2019 +0530
 9732 
 9733     cluster/ta: Notify the clients only if there are pending heals
 9734     
 9735     Problem:
 9736     In case of thin arbiter, before index healer starts crawling the
 9737     indices at every heal-timeout interval, even if there is nothing to
 9738     be healed it will send an upcall notification to all the clients to
 9739     release any AFR_TA_DOM_NOTIFY locks that they hold. SHD will wait
 9740     for the upcall to return before proceeding with the heal even though
 9741     there is nothing to be healed. This will also invalidates the cached
 9742     information about the bricks states on the clients which leads to
 9743     extra calls on TA from clients for the next reads & writes if needed.
 9744     This will impact the IO performance.
 9745     
 9746     Fix:
 9747     - Before sending the upcall to the clients, check for any pending heals
 9748     on TA without taking  any locks.
 9749     - If there is nothing marked bad on TA, then continue with the index
 9750     crawl to heal any dirty markings present on the files due to any post-op
 9751     failure.
 9752     - If there is a brick marked as bad on TA, then take the
 9753     AFR_TA_DOM_NOTIFY lock on TA from SHD, get the state on TA and
 9754     continue with the current healing process.
 9755     
 9756     Change-Id: Ieb477bc6cb18bbdfd4e7a0453c5ed79b574ec9d6
 9757     fixes: bz#1724184
 9758     Signed-off-by: karthik-us <ksubrahm@redhat.com>
 9759 
 9760 commit 1a38a6f4853f87ac4e45240579581292290f4f27
 9761 Author: karthik-us <ksubrahm@redhat.com>
 9762 Date:   Wed Jun 12 11:57:02 2019 +0530
 9763 
 9764     cluster/afr: Fix incorrect reporting of gfid & type mismatch
 9765     
 9766     Problems:
 9767     1. When checking for type and gfid mismatch, if the type or gfid
 9768     is unknown because of missing gfid handle and the gfid xattr
 9769     it will be reported as type or gfid mismatch and the heal will
 9770     not complete.
 9771     
 9772     2. If the source selected during entry heal has null gfid the same
 9773     will be sent to afr_lookup_and_heal_gfid(). In this function when
 9774     we try to assign the gfid on the bricks where it does not exist,
 9775     we are considering the same gfid and try to assign that on those
 9776     bricks. This will fail in posix_gfid_set() since the gfid sent
 9777     is null.
 9778     
 9779     Fix:
 9780     If the gfid sent to afr_lookup_and_heal_gfid() is null choose a
 9781     valid gfid before proceeding to assign the gfid on the bricks
 9782     where it is missing.
 9783     
 9784     In afr_selfheal_detect_gfid_and_type_mismatch(), do not report
 9785     type/gfid mismatch if the type/gfid is unknown or not set.
 9786     
 9787     Change-Id: Ia06552e4dc4a9f89cb7f5302833604bd21bbf7da
 9788     fixes: bz#1722507
 9789     Signed-off-by: karthik-us <ksubrahm@redhat.com>
 9790 
 9791 commit 3c1efa0c92445638bbfa57c2e868c79f7d987fc3
 9792 Author: Vijay Bellur <vbellur@redhat.com>
 9793 Date:   Thu Jun 20 15:50:01 2019 -0700
 9794 
 9795     Replace usleep() with nanosleep()
 9796     
 9797     As usleep has been obsoleted, changed all invocations of usleep
 9798     to nanosleep. From man 3 usleep:
 9799     
 9800     "4.3BSD, POSIX.1-2001.  POSIX.1-2001 declares this function
 9801      obsolete; use nanosleep(2) instead. POSIX.1-2008 removes the
 9802      specification of usleep()."
 9803     
 9804     Added a helper function gf_nanosleep() to have a single place
 9805     for handling edge cases that might arise from the conversion of
 9806     usleep to nanosleep and allow the sleep to resume with right
 9807     remaining value upon being interrupted.
 9808     
 9809     Fixes: bz#1721686
 9810     
 9811     Change-Id: Ia39ab82c9e0f4669d2c00d4cdf25e38d94ef9f62
 9812     Signed-off-by: Vijay Bellur <vbellur@redhat.com>
 9813 
 9814 commit b82743a5822626e83b233956803421fda9b1c3f9
 9815 Author: Vijay Bellur <vbellur@redhat.com>
 9816 Date:   Tue Jul 9 16:32:36 2019 -0700
 9817 
 9818     Remove hadoop related code from the codebase
 9819     
 9820     As Hadoop is no longer supported, dropping code for
 9821     handling Hadoop access.
 9822     
 9823     Fixes: bz#1728417
 9824     Signed-off-by: Vijay Bellur <vbellur@redhat.com>
 9825     Change-Id: I8fcf4faacb364f1c9a8abb0c48faec337087f845
 9826 
 9827 commit ad828d2e0c930b31b870c92fbaeac7a43fee08da
 9828 Author: kshithijiyer <kshithij.ki@gmail.com>
 9829 Date:   Thu Jul 4 19:03:34 2019 +0530
 9830 
 9831     [geo-replication] geo-replication/setup.py missing license details in setup()
 9832     
 9833     geo-replication/setup.py is missing license details in
 9834     setup() and is using string 'syncdaemon' instead
 9835     of using the variable name which has the same value.
 9836     
 9837     Present:
 9838     setup(
 9839         name=name,
 9840         version="",
 9841         description='GlusterFS Geo Replication',
 9842         license='',
 9843         author='Red Hat, Inc.',
 9844         author_email='gluster-devel@gluster.org',
 9845         url='http://www.gluster.org',
 9846         packages=['syncdaemon', ],
 9847         test_suite='nose.collector',
 9848         install_requires=[],
 9849         scripts=[],
 9850         entry_points={},
 9851     )
 9852     
 9853     Changing to:
 9854     setup(
 9855         name=name,
 9856         version="",
 9857         description='GlusterFS Geo Replication',
 9858         license='GPLV2 and LGPLV3+',
 9859         author='Red Hat, Inc.',
 9860         author_email='gluster-devel@gluster.org',
 9861         url='http://www.gluster.org',
 9862         packages=[name, ],
 9863         test_suite='nose.collector',
 9864         install_requires=[],
 9865         scripts=[],
 9866         entry_points={},
 9867     )
 9868     
 9869     fixes: bz#1727107
 9870     
 9871     Change-Id: I742c4c510da55a22256c819da7e2fc9622119f63
 9872     Signed-off-by: kshithijiyer <kshithij.ki@gmail.com>
 9873 
 9874 commit c2e514c18df8fe1dfe06bc147f94756838f67c41
 9875 Author: Amar Tumballi <amarts@redhat.com>
 9876 Date:   Sun May 5 13:19:31 2019 +0530
 9877 
 9878     rpc/xdr: include nfs specific files in build only if gNFS is enabled
 9879     
 9880     updates: bz#1193929
 9881     Change-Id: I2b85fd0a04c77815a154f445ec8fb4da37dcbe40
 9882     Signed-off-by: Amar Tumballi <amarts@redhat.com>
 9883 
 9884 commit 670299cf4fc90c1c61032057b6ff570fee7a448e
 9885 Author: Mohammed Rafi KC <rkavunga@redhat.com>
 9886 Date:   Mon Jun 24 12:00:20 2019 +0530
 9887 
 9888     glusterd/svc: update pid of mux volumes from the shd process
 9889     
 9890     For a normal volume, we are updating the pid from a the
 9891     process while we do a daemonization or at the end of the
 9892     init if it is no-daemon mode. Along with updating the pid
 9893     we also lock the file, to make sure that the process is
 9894     running fine.
 9895     
 9896     With brick mux, we were updating the pidfile from gluterd
 9897     after an attach/detach request.
 9898     
 9899     There are two problems with this approach.
 9900     1) We are not holding a pidlock for any file other than parent
 9901        process.
 9902     2) There is a chance for possible race conditions with attach/detach.
 9903        For example, shd start and a volume stop could race. Let's say
 9904        we are starting an shd and it is attached to a volume.
 9905        While we trying to link the pid file to the running process,
 9906        this would have deleted by the thread that doing a volume stop.
 9907     
 9908     Change-Id: I29a00352102877ce09ea3f376ca52affceb5cf1a
 9909     Updates: bz#1722541
 9910     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
 9911 
 9912 commit a94d7433169cbf29b9109c96ac57b99e898e6f17
 9913 Author: Sheetal Pamecha <spamecha@redhat.com>
 9914 Date:   Mon Jul 8 15:09:39 2019 +0530
 9915 
 9916     posix: fix Wformat-overflow warning
 9917     
 9918     warning: ‘%s’ directive argument is null
 9919     
 9920     Change-Id: I2ce9560f98a8310886c31384e40c2e101ad2c719
 9921     updates: bz#1193929
 9922     Signed-off-by: Sheetal Pamecha <spamecha@redhat.com>
 9923 
 9924 commit d0d77953ed4020a8c7ac87073bd5e6c4743bc961
 9925 Author: Sheetal Pamecha <spamecha@redhat.com>
 9926 Date:   Fri Jul 5 16:51:45 2019 +0530
 9927 
 9928     glusterd: fix packed member address warning
 9929     
 9930     warning: taking address of packed member of
 9931     ‘struct _quota_limits’ may result in an
 9932     unaligned pointer value.
 9933     
 9934     Change-Id: Ib889c99184560f0d24dbcc4fae10b563a2b9b7fd
 9935     updates: bz#1193929
 9936     Signed-off-by: Sheetal Pamecha <spamecha@redhat.com>
 9937 
 9938 commit 4c44c47a40910c70e7d22d2bc6b7c21b88e3296b
 9939 Author: Atin Mukherjee <amukherj@redhat.com>
 9940 Date:   Tue Apr 9 19:02:53 2019 +0530
 9941 
 9942     quick-read: rename cache-invalidation key to avoid redundant keys
 9943     
 9944     With group-metadata-cache group profile settings
 9945     performance.cache-invalidation option when turned on enables both md-cache
 9946     and quick-read xlator's cache-invalidation feature. While the intent of
 9947     the group-metadata-cache is to set md-cache xlator's cache-invalidation
 9948     feature, quick-read xlator also gets affected due to the same. While
 9949     md-cache feature and it's profile existed since release-3.9,
 9950     quick-read cache-invalidation was introduced in release-4 and due to
 9951     this op-version mismatch on any cluster which is >= glusterfs-4 when
 9952     this group profile is applied it breaks backward compatibility with the
 9953     old clients.
 9954     
 9955     The proposed fix here is to rename the key in quick-read to
 9956     'quick-read-cache-invalidation' so that both these features have
 9957     distinct identification. While this brings in by itself a backward
 9958     compatibility challenge where this feature is enabled in an existing
 9959     cluster and when the same is upgraded to a version where this change
 9960     exists, it will lead to an unidentified old key. But as a workaround we
 9961     can always ask users upgrading to release-7 version to turn off this
 9962     option, upgrade the cluster and turn it back on with the new key. This
 9963     needs to be documented once the patch is accepted.
 9964     
 9965     Fixes: bz#1698042
 9966     Change-Id: I30422ba6496208e21191a8d78ad29b2e21078664
 9967     Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
 9968     Signed-off-by: Raghavendra G <rgowdapp@redhat.com>
 9969 
 9970 commit 99d210a704d2e85c95fac5edcf435bd059aad368
 9971 Author: Xie Changlong <xiechanglong@cmss.chinamobile.com>
 9972 Date:   Fri Jul 5 15:16:23 2019 +0800
 9973 
 9974     gnfs: use strcpy to prevent memory overflow
 9975     
 9976     fixes: bz#1727248
 9977     Change-Id: Iea289032a8feecf2945668d3fb44a6a53089fdea
 9978     Signed-off-by: Xie Changlong <xiechanglong@cmss.chinamobile.com>
 9979 
 9980 commit e21544483c25908e9ed142480e9753d646d53605
 9981 Author: Yaniv Kaul <ykaul@redhat.com>
 9982 Date:   Fri Jun 21 14:24:42 2019 +0300
 9983 
 9984     glusterd.h: align structs
 9985     
 9986     Move some variables around to reduce compiler padding.
 9987     
 9988     Change-Id: I9cd53ce257fb169270c295ac60d58d4a6a950d4f
 9989     updates: bz#1193929
 9990     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
 9991 
 9992 commit af989db23d1db00e087f2b9d3dfc43b13ef17153
 9993 Author: Mohit Agrawal <moagrawal@redhat.com>
 9994 Date:   Wed Jul 3 15:22:38 2019 +0530
 9995 
 9996     glusterd: Show the correct brick status in get-state
 9997     
 9998     Problem: get-state does not show correct brick status if brick
 9999              status is not Started, it always shows started if any value
10000              is set brickinfo->status
10001     
10002     Solution: Check the value of brickinfo->status to show correct status
10003               in get-state
10004     
10005     Change-Id: I12a79619024c2cf59f338220d144f2f034059b3b
10006     fixes: bz#1726906
10007     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
10008 
10009 commit 7d270d05f835d41e32572501246b50181dc9be56
10010 Author: Sanju Rakonde <srakonde@redhat.com>
10011 Date:   Tue Jul 2 15:23:00 2019 +0530
10012 
10013     glusterd: don't log a warning message for tier-enabled key
10014     
10015     We are logging a warning message saying unknown-key for
10016     tier-enabled kay. although the tier xlator is deprecated,
10017     this key is left behind for handling the peer rejection
10018     issues in a heterogeneous cluster. We need not to log if
10019     this key is not found/recognised.
10020     
10021     updates: bz#1193929
10022     
10023     Change-Id: Ia68661898a618f99a240ca8d8a124ff6a65ebe9d
10024     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
10025 
10026 commit 36f6e6df0ff15d0464b869803710adca2b65e8ba
10027 Author: Raghavendra Bhat <raghavendra@redhat.com>
10028 Date:   Tue Jun 25 10:51:33 2019 -0400
10029 
10030     features/snapview-server: use the same volfile server for gfapi options
10031     
10032     snapview server xlator makes use of "localhost" as the volfile server while
10033     initing the new glfs instance to talk to a snapshot. While localhost is fine,
10034     better use the same volfile server that was used to start the snapshot
10035     daemon containing the snapview-server xlator.
10036     
10037     Change-Id: I4485d39b0e3d066f481adc6958ace53ea33237f7
10038     fixes: bz#1725211
10039     Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
10040 
10041 commit 532b0fc8b1ace9ad48fdaf643e0b1a34020b6cd8
10042 Author: N Balachandran <nbalacha@redhat.com>
10043 Date:   Fri Jun 21 09:04:19 2019 +0530
10044 
10045     cluster/dht:  Fixed a memleak in dht_rename_cbk
10046     
10047     Fixed a memleak in dht_rename_cbk when creating
10048     a linkto file.
10049     
10050     Change-Id: I705adef3cb79e33806520fc2b15558e90e2c211c
10051     fixes: bz#1722698
10052     Signed-off-by: N Balachandran <nbalacha@redhat.com>
10053 
10054 commit 080827829eabd3cbed1a7feaa37de091834fcbb1
10055 Author: Sheetal Pamecha <spamecha@redhat.com>
10056 Date:   Mon Jul 1 20:48:31 2019 +0530
10057 
10058     cli: Remove Wformat-truncation compiler warning
10059     
10060     This warning is issued due to unhandled output truncation.
10061     As in the code, truncation is expected, we can remove this
10062     warning by placing a check on the return value of the function
10063     and handling it.
10064     In this way, the warning will not be issued.
10065     
10066     Change-Id: I1820b58fe9a7601961c20944b259df322db35057
10067     updates: bz#1193929
10068     Signed-off-by: Sheetal Pamecha <spamecha@redhat.com>
10069 
10070 commit cbc234d9ddfb7e2f0089d7ba9d1dd81b0367ba8f
10071 Author: Amar Tumballi <amarts@redhat.com>
10072 Date:   Mon Jun 24 19:24:05 2019 +0530
10073 
10074     glusterfs-fops: fix the modularity
10075     
10076     glusterfs-fops.h was moved to rpc/xdr to support compound fops.
10077     (ref: https://review.gluster.org/14032, 2f945b86d3)
10078     
10079     This was fine as long as all these header files were in single
10080     include directory after 'install'. With the move to separate out
10081     glusterfs specific header files into another directory inside
10082     /usr/include (ref: https://review.gluster.org/21746, 20ef211cfa),
10083     glusterfs-fops.h file was not in the proper path when an external
10084     .c file tried to include any of glusterfs specific .h file (like
10085     xlator.h).
10086     
10087     Now, we have removed compound-fops, with that, none of the enums
10088     declared in glusterfs-fops.h are actually getting used on wire
10089     anymore. Hence, it makes sense to get this to libglusterfs/src
10090     as a single point of definition. With this change, the external
10091     programs can use glusterfs header files.
10092     
10093     also remove some enum definitions which are not used in code
10094     anymore.
10095     
10096     Updates: bz#1636297
10097     Change-Id: I423c44d3dbe2efc777299c544ece3cb172fc7e44
10098     Signed-off-by: Amar Tumballi <amarts@redhat.com>
10099 
10100 commit c72b3f13452fcd9289230516f67d7c07951a8d5c
10101 Author: kshithijiyer <kshithij.ki@gmail.com>
10102 Date:   Fri Jun 28 15:32:31 2019 +0530
10103 
10104     Removing one top command from gluster v help
10105     
10106     The current help show 2 different top commands
10107     intead of one single top command which can be
10108     easily observed when "# gluster v help" command
10109     is issued. Removing one "volume top <VOLNAME>"
10110     and clubbing into them into a single command.
10111     
10112     Current help:
10113     volume top <VOLNAME> {open|read|write|opendir|readdir|clear}
10114     [nfs|brick <brick>] [list-cnt <value>] |
10115     volume top <VOLNAME> {read-perf|write-perf}
10116     [bs <size> count <count>] [brick <brick>]
10117     [list-cnt <value>] - volume top operations
10118     
10119     Expected help:
10120     volume top <VOLNAME> {open|read|write|opendir|readdir|clear}
10121     [nfs|brick <brick>] [list-cnt <value>] | {read-perf|write-perf}
10122     [bs <size> count <count>] [brick <brick>] [list-cnt <value>]
10123     - volume top operations
10124     
10125     fixes: bz#1725034
10126     
10127     Change-Id: Ifbc4c95f2558286e27dfc5e9667046b80eb1715d
10128     Signed-off-by: kshithijiyer <kshithij.ki@gmail.com>
10129 
10130 commit a194e4dc78c50605e5178e56976227ccefb8cdb1
10131 Author: Sanju Rakonde <srakonde@redhat.com>
10132 Date:   Mon Jun 17 17:54:01 2019 +0530
10133 
10134     core: use multiple servers while mounting a volume using ipv6
10135     
10136     According to man page:
10137     mount -t glusterfs [-o <options>] <server1>,<server2>,
10138                   <server3>,..<serverN>:/<volname>[/<subdir>] <mount_point>
10139     
10140     When we try
10141     
10142     mount -t glusterfs 52:54:00:23:5a:b6,52:54:00:a0:be:ed:/ta-vol1 /mnt/ta
10143     
10144     we see the below msg in log:
10145     
10146     [2019-06-17 11:41:36.085809] I [MSGID: 100030] [glusterfsd.c:2867:main] 0-/usr/local/sbin/glusterfs: Started running /usr/local/sbin/glusterfs version 7dev (args: /usr/local/sbin/glusterfs --process-name fuse --volfile-server=52:54:00:23:5a --volfile-id=/ta-vol1 /mnt/ta)
10147     
10148     With this change, we'll able to give multiple volfile servers
10149     while mounting the volume using ipv6.
10150     
10151     After the change, I see the following in log:
10152     
10153     [2019-06-17 12:00:21.183658] I [MSGID: 100030] [glusterfsd.c:2867:main] 0-/usr/local/sbin/glusterfs: Started running /usr/local/sbin/glusterfs version 7dev (args: /usr/local/sbin/glusterfs --process-name fuse --volfile-server=52:54:00:23:5a:b6 --volfile-server=52:54:00:a0:be:ed --volfile-id=/ta-vol1 /mnt/ta)
10154     
10155     fixes: bz#1719290
10156     
10157     credits: Aga <aga_1990@hotmail.com>
10158     
10159     Change-Id: Icf89bea3ba15d8374ef428aeb59f2ef55ad544ec
10160     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
10161 
10162 commit 6fbf323d5f72db879d733af5ba9c67c7a1bfec63
10163 Author: Atin Mukherjee <amukherj@redhat.com>
10164 Date:   Thu Jun 27 10:49:30 2019 +0530
10165 
10166     glusterd: fix clang scan defects
10167     
10168     Fixes following:
10169     https://build.gluster.org/job/clang-scan/744/clangScanBuildBugs/browse/report-dd8d31.html#EndPath
10170     https://build.gluster.org/job/clang-scan/744/clangScanBuildBugs/browse/report-89a1cd.html#EndPath
10171     
10172     Updates: bz#1622665
10173     Change-Id: Ibd201fb2ca54ae7ae3fed8a8d87815358b614349
10174     Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
10175 
10176 commit 22e2d990dbbea878dba5569cc2d0df33955913b1
10177 Author: Xavi Hernandez <xhernandez@redhat.com>
10178 Date:   Wed Jun 26 16:23:15 2019 +0200
10179 
10180     rfc.sh: Improve bug identification
10181     
10182     When a backport is done and there is more than one bug reference in the
10183     commit message (normally we include the old bug reference in the commit
10184     message), old rfc.sh may choose the incorrect bug id.
10185     
10186     This patch improves the regular expression to locate the bug id.
10187     
10188     Updates: bz#1193929
10189     Change-Id: Ifaa4e50c0ddba21c16ef0c5a1f6d46c315cf3378
10190     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
10191 
10192 commit 9b223b15ab69fce4076de036ee162f36a058bcd2
10193 Author: Vishal Pandey <vpandey@redhat.com>
10194 Date:   Wed Apr 24 13:37:16 2019 +0530
10195 
10196     glusterd/thin-arbiter: Thin-arbiter integration with GD1
10197     
10198     gluster volume create <VOLNAME> replica 2 thin-arbiter 1 <host1>:<brick1> <host2>:<brick2>
10199     <thin-arbiter-host>:<path-to-store-replica-id-file> [force]
10200     
10201     The changes have been made in a way that the last brick in the bricks list
10202     will be treated as the thin-arbiter.
10203     GD1 will be manipulated to consider replica count to be as 2 and continue creating the
10204     volume like any other replica 2 volume but since thin-arbiter volumes need ta-brick
10205     client xlator entries for each subvolume in fuse volfile, volfile generation is
10206     modified in a way to inject these entries seperately in the volfile for every subvolume.
10207     
10208     Few more additions -
10209     1- Save the volinfo with new fields ta_bricks list and thin_arbiter_count.
10210     2- Introduce a new option client.ta-brick-port to add remote-port to ta-brick xlator entry
10211        in fuse volfiles. The option can be set using the following CLI syntax -
10212        gluster volume set <VOLNAME> client.ta-brick-port <PORTNO.>
10213     3- Volume Info will contain a Thin-Arbiter-path entry to distinguish
10214        from other replicate volumes.
10215     
10216     Change-Id: Ib434e2313b29716f32476c6c211d282c4ef39406
10217     Updates #687
10218     Signed-off-by: Vishal Pandey <vpandey@redhat.com>
10219 
10220 commit 29ad22aa9482a69f3fcf04eea762e76602bbe9a0
10221 Author: Sunny Kumar <sunkumar@redhat.com>
10222 Date:   Thu Jun 20 18:47:45 2019 +0530
10223 
10224     geo-rep : fix gluster command path for non-root session
10225     
10226     Problem:
10227     gluster command not found.
10228     
10229     Cause:
10230     In Volinfo class we issue command 'gluster vol info' to get information
10231     about volume like getting brick_root to perform various operation.
10232     When geo-rep session is configured for non-root user Volinfo class
10233     fails to issue gluster command due to unavailability of gluster
10234     binary path for non-root user.
10235     
10236     Solution:
10237     Use config value 'slave-gluster-command-dir'/'gluster-command-dir' to get path
10238     for gluster command based on caller.
10239     
10240     fixes: bz#1722740
10241     
10242     Change-Id: I4ec46373da01f5d00ecd160c4e8c6239da8b3859
10243     Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
10244 
10245 commit d4a4c4284317f3a51e2edfc218bfd030103f1110
10246 Author: Xavi Hernandez <xhernandez@redhat.com>
10247 Date:   Wed Jun 26 17:34:23 2019 +0200
10248 
10249     run-tests.sh: stop test on Ctrl-C
10250     
10251     When a test is run through 'timeout' command, Ctrl-C becomes useless and
10252     the test cannot be stopped.
10253     
10254     Added '--foreground' option to 'timeout' command that makes it possible
10255     to use Ctrl-C.
10256     
10257     Updates: bz#1193929
10258     Change-Id: I160984d1aba470e3a4a1135bd5fa57471046f52f
10259     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
10260 
10261 commit efa8477013d98a26283302ecb42d4ba606ce405e
10262 Author: Mohammed Rafi KC <rkavunga@redhat.com>
10263 Date:   Thu Jun 27 19:17:29 2019 +0530
10264 
10265     graph/shd: Use top down approach while cleaning xlator
10266     
10267     We were cleaning xlator from botton to top, which might
10268     lead to problems when upper xlators trying to access
10269     the xlator object loaded below.
10270     
10271     One such scenario is when fd_unref happens as part of the
10272     fini call which might lead to calling the releasedir to
10273     lower xlator. This will lead to invalid mem access
10274     
10275     Change-Id: I8a6cb619256fab0b0c01a2d564fc88287c4415a0
10276     Updates: bz#1716695
10277     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
10278 
10279 commit 5d66eafec581fb3209af74595784be8854ca40a4
10280 Author: Raghavendra G <rgowdapp@redhat.com>
10281 Date:   Tue Jun 4 19:22:45 2019 +0530
10282 
10283     protocol/client: propagte GF_EVENT_CHILD_PING only for connections to brick
10284     
10285     Two reasons:
10286     * ping responses from glusterd may not be relevant for Halo
10287       replication. Instead, it might be interested in only knowing whether
10288       the brick itself is responsive.
10289     * When a brick is killed, propagating GF_EVENT_CHILD_PING of ping
10290       response from glusterd results in GF_EVENT_DISCONNECT spuriously
10291       propagated to parent xlators. These DISCONNECT events are from the
10292       connections client establishes with glusterd as part of its
10293       reconnect logic. Without GF_EVENT_CHILD_PING, the last event
10294       propagated to parent xlators would be the first DISCONNECT event
10295       from brick and hence subsequent DISCONNECTS to glusterd are not
10296       propagated as protocol/client prevents same event being propagated
10297       to parent xlators consecutively. propagating GF_EVENT_CHILD_PING for
10298       ping responses from glusterd would change the last_sent_event to
10299       GF_EVENT_CHILD_PING and hence protocol/client cannot prevent
10300       subsequent DISCONNECT events
10301     
10302     Signed-off-by: Raghavendra G <rgowdapp@redhat.com>
10303     Fixes: bz#1716979
10304     Change-Id: I50276680c52f05ca9e12149a3094923622d6eaef
10305 
10306 commit 7d8be567f2f904fc74d0990ebce2e8afbedab918
10307 Author: Jiffin Tony Thottan <jthottan@redhat.com>
10308 Date:   Wed Jun 26 15:58:33 2019 +0530
10309 
10310     posix : add posix_set_ctime() in posix_ftruncate()
10311     
10312     Change-Id: I0cb5320fea71306e0283509ae47024f23874b53b
10313     fixes: bz#1723761
10314     Signed-off-by: Jiffin Tony Thottan <jthottan@redhat.com>
10315 
10316 commit e8f8d16fc4b8e5be48f4d7c9ff9d170934ffb7fc
10317 Author: Mohammed Rafi KC <rkavunga@redhat.com>
10318 Date:   Mon Jun 24 15:49:04 2019 +0530
10319 
10320     graph/shd: Use glusterfs_graph_deactivate to free the xl rec
10321     
10322     We were using glusterfs_graph_fini to free the xl rec from
10323     glusterfs_process_volfp as well as glusterfs_graph_cleanup.
10324     
10325     Instead we can use glusterfs_graph_deactivate, which is does
10326     fini as well as other common rec free.
10327     
10328     Change-Id: Ie4a5f2771e5254aa5ed9f00c3672a6d2cc8e4bc1
10329     Updates: bz#1716695
10330     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
10331 
10332 commit f782881f825efcd4ab3b74209c1c7aca3796023b
10333 Author: Shwetha K Acharya <sacharya@redhat.com>
10334 Date:   Wed May 29 16:49:01 2019 +0530
10335 
10336     geo-rep: Upgrading config file to new version
10337     
10338     - configuration handling is enhanced with patch
10339     https://review.gluster.org/#/c/glusterfs/+/18257/
10340     - hence, the old configurations are not applied when
10341     Geo-rep session is created in the old version and upgraded.
10342     
10343     This patch solves the issue. It,
10344     - checks if the config file is old.
10345     - parses required values from old config file and stores in new
10346       config file, which ensures that configerations are applied on
10347       upgrade.
10348     - stores old config file as backup.
10349     - handles changes in options introduced in
10350       https://review.gluster.org/#/c/glusterfs/+/18257/
10351     
10352     fixes: bz#1707731
10353     Change-Id: Iad8da6c1e1ae8ecf7c84dfdf8ea3ac6966d8a2a0
10354     Signed-off-by: Shwetha K Acharya <sacharya@redhat.com>
10355 
10356 commit 5cbc87d8b8f1287e81c38b793b8d13b057208c62
10357 Author: Sheetal Pamecha <spamecha@redhat.com>
10358 Date:   Wed Jun 19 15:08:58 2019 +0530
10359 
10360     posix: modify storage.reserve option to take size and percent
10361     
10362     * reverting changes made in
10363     https://review.gluster.org/#/c/glusterfs/+/21686/
10364     
10365     * Now storage.reserve can take value in percent or bytes
10366     
10367     fixes: bz#1651445
10368     Change-Id: Id4826210ec27991c55b17d1fecd90356bff3e036
10369     Signed-off-by: Sheetal Pamecha <spamecha@redhat.com>
10370 
10371 commit 1716a907da1a835b658740f1325033d7ddd44952
10372 Author: Xavi Hernandez <xhernandez@redhat.com>
10373 Date:   Fri Jun 21 11:28:08 2019 +0200
10374 
10375     core: fix memory allocation issues
10376     
10377     Two problems have been identified that caused that gluster's memory
10378     usage were twice higher than required.
10379     
10380     1. An off by 1 error caused that all objects allocated from the memory
10381        pools were taken from a pool bigger than required. Since each pool
10382        corresponds to a size equal to a power of two, this was wasting half
10383        of the available memory.
10384     
10385     2. The header information used for accounting on each memory object was
10386        not taken into consideration when searching for a suitable memory
10387        pool. It was added later when each individual block was allocated.
10388        This made this space "invisible" to memory accounting.
10389     
10390     Credits: Thanks to Nithya Balachandran for identifying this problem and
10391              testing this patch.
10392     
10393     Fixes: bz#1722802
10394     Change-Id: I90e27ad795fe51ca11c13080f62207451f6c138c
10395     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
10396 
10397 commit dc1b87fcfef08c9497b0c02b2410c9d18bbc2dba
10398 Author: Michael Adam <obnox@samba.org>
10399 Date:   Thu Jun 20 13:09:37 2019 +0200
10400 
10401     [RFC] change get_real_filename implementation to use ENOATTR instead of ENOENT
10402     
10403     get_real_filename is implemented as a virtual extended attribute to help
10404     Samba implement the case-insensitive but case preserving SMB protocol
10405     more efficiently. It is implemented as a getxattr call on the parent directory
10406     with the virtual key of "get_real_filename:<entryname>" by looking for a
10407     spelling with different case for the provided file/dir name (<entryname>)
10408     and returning this correct spelling as a result if the entry is found.
10409     Originally (05aaec645a6262d431486eb5ac7cd702646cfcfb), the
10410     implementation used the ENOENT errno to return the authoritative answer
10411     that <entryname> does not exist in any case folding.
10412     
10413     Now this implementation is actually a violation or misuse of the defined
10414     API for the getxattr call which returns ENOENT for the case that the dir
10415     that the call is made against does not exist and ENOATTR (or the synonym
10416     ENODATA) for the case that the xattr does not exist.
10417     
10418     This was not a problem until the gluster fuse-bridge was changed
10419     to do map ENOENT to ESTALE in 59629f1da9dca670d5dcc6425f7f89b3e96b46bf,
10420     after which we the getxattr call for get_real_filename returned an
10421     ESTALE instead of ENOENT breaking the expectation in Samba.
10422     
10423     It is an independent problem that ESTALE should not leak out to user
10424     space but is intended to trigger retries between fuse and gluster.
10425     But nevertheless, the semantics seem to be incorrect here and should
10426     be changed.
10427     
10428     This patch changes the implementation of the get_real_filename virtual
10429     xattr to correctly return ENOATTR instead of ENOENT if the file/directory
10430     being looked up is not found.
10431     
10432     The Samba glusterfs_fuse vfs module which takes advantage of the
10433     get_real_filename over a fuse mount will receive a corresponding change
10434     to map ENOATTR to ENOENT. Without this change, it will still work
10435     correctly, but the performance optimization for nonexisting files is
10436     lost. On the other hand side, this change removes the distinction
10437     between the old not-implemented case and the implemented case.
10438     So Samba changed to treat ENOATTR like ENOENT will not work correctly
10439     any more against old servers that don't implement get_real_filename.
10440     I.e. existing files will be reported as non-existing
10441     
10442     Change-Id: I971b427ab8410636d5d201157d9af70e0d075b67
10443     fixes: bz#1722977
10444     Signed-off-by: Michael Adam <obnox@samba.org>
10445 
10446 commit cfc30c15ba54fa91cfa8c62b961553b416bb9ff7
10447 Author: Xavi Hernandez <xhernandez@redhat.com>
10448 Date:   Thu Jun 6 08:12:34 2019 +0200
10449 
10450     locks: enable notify-contention by default
10451     
10452     This patch enables the lock contention notification by default.
10453     
10454     Change-Id: I10131b026a7cb09fc7c93e1e6c8549988c1d7751
10455     Fixes: bz#1717754
10456     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
10457 
10458 commit 323389858562f28f16ec2e32e0160140f7d76668
10459 Author: Raghavendra Bhat <raghavendra@redhat.com>
10460 Date:   Thu Jun 20 11:19:47 2019 -0400
10461 
10462     statedump: dump the minimum and maximum latency seen by each xlator
10463     
10464     Change-Id: Ie9661282694eb9dcfb0d0bc6af512610e02f7347
10465     fixes: bz#1722598
10466     Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
10467 
10468 commit f36086db87aae24c10abde434f081d78b942735e
10469 Author: Xavi Hernandez <xhernandez@redhat.com>
10470 Date:   Tue Jun 25 18:00:06 2019 +0200
10471 
10472     glusterd: fix use-after-free of a dict_t
10473     
10474     A dict was passed to a function that calls dict_unref() without taking
10475     any additional reference. Given that the same dict is also used after
10476     the function returns, this was causing a use-after-free situation.
10477     
10478     To fix the issue, we simply take an additional reference before calling
10479     the function.
10480     
10481     Fixes: bz#1723890
10482     Change-Id: I98c6b76b08fe3fa6224edf281a26e9ba1ffe3017
10483     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
10484 
10485 commit 40c18b189f31a87b5c5f9344c4a0ab44cb61ef75
10486 Author: Kinglong Mee <mijinlong@horiscale.com>
10487 Date:   Mon May 13 16:42:01 2019 +0800
10488 
10489     gfapi: set right pargfid according to parent's inode
10490     
10491     EC updates parent's inode from loc, when pargfid is not the parent's inode's,
10492     a warnning message is print, and an error is returned to up xlator.
10493     
10494     [ec-helpers.c:400:ec_loc_gfid_check] 0-openfs1-disperse-0: Mismatching GFID's in loc
10495     
10496     Change-Id: Ie8036be91f039083a1e3fe17ec235a222df84b97
10497     fixes: bz#1723280
10498     Signed-off-by: Kinglong Mee <mijinlong@horiscale.com>
10499 
10500 commit d58594d802d617ed81b8f991fd312c224f1f85d4
10501 Author: Raghavendra Bhat <raghavendra@redhat.com>
10502 Date:   Tue Apr 30 15:38:55 2019 -0400
10503 
10504     * core: do not assert in inode_unref if the inode table cleanup has started
10505     
10506     There is a good chance that, the inode on which unref came has already been
10507     zero refed and added to the purge list. This can happen when inode table is
10508     being destroyed (glfs_fini is something which destroys the inode table).
10509     
10510     Consider a directory 'a' which has a file 'b'. Now as part of inode table destruction
10511     zero refing of inodes does not happen from leaf to the root. It happens in the order
10512     inodes are present in the list. So, in this example, the dentry of 'b' would have its
10513     parent set to the inode of 'a'. So if 'a' gets zero refed first (as part of
10514     inode table cleanup) and then 'b' has to zero refed, then dentry_unset is called on
10515     the dentry of 'b' and it further goes on to call inode_unref on b's parent which is
10516     'a'. In this situation, GF_ASSERT would be called as the refcount of 'a' has been
10517     already set to zero.
10518     
10519     So, return the inode (in the function inode_unref without doing anything) if the
10520     inode table cleanup has already started and inode's refcount is zero.
10521     
10522     Change-Id: I91e0a807d5c9ce0daae5a611c38da379fd11076e
10523     fixes: bz#1722546
10524     Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
10525 
10526 commit bee5d7bd8a602184c94c12115c7e30572eb7588c
10527 Author: kshithijiyer <kshithij.ki@gmail.com>
10528 Date:   Mon Jun 24 20:08:48 2019 +0530
10529 
10530     Adding white spaces to description of set group.
10531     
10532     The description of set group is missing spaces which
10533     leads to the description look like:
10534     volume set <VOLNAME> group  <GROUP> - This option can be used for
10535     setting multiple pre-defined volume optionswhere group_name is a
10536     file under /var/lib/glusterd/groups containing onekey, value pair
10537     per line
10538     
10539     Instead of:
10540     volume set <VOLNAME> group <GROUP> - This option can be used for
10541     setting multiple pre-defined volume options where group_name is a
10542     file under /var/lib/glusterd/groups containing one key value
10543     pair per line
10544     
10545     Fixes: bz#1723455
10546     
10547     Change-Id: I4957988c0c1f35f043db3f64089c049193e60e8f
10548     Signed-off-by: kshithijiyer <kshithij.ki@gmail.com>
10549 
10550 commit a72452fcf90679b28baec12d2769cbaa982bb4e4
10551 Author: Atin Mukherjee <amukherj@redhat.com>
10552 Date:   Tue Jun 25 11:11:10 2019 +0530
10553 
10554     glusterd: conditionally clear txn_opinfo in stage op
10555     
10556     ...otherwise this leads to a crash when volume status is run on a
10557     heterogeneous mode.
10558     
10559     Fixes: bz#1723658
10560     Change-Id: I0d39f412b2e5e9d3ef0a3462b90b38bb5364b09d
10561     Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
10562 
10563 commit 2d7b77eb971700c1073db2b74f5877c1ae8293fc
10564 Author: Mohit Agrawal <moagrawal@redhat.com>
10565 Date:   Mon Jun 24 10:57:49 2019 +0530
10566 
10567     glusterd: Can't run rebalance due to long unix socket
10568     
10569     Problem: glusterd populate unix socket file name based
10570              on volname and if volname is lengthy socket
10571              system call's are failed due to breach maximum length
10572              is defined in the kernel.
10573     
10574     Solution:Convert unix socket name to hash to resolve the issue
10575     
10576     Change-Id: I5072e8184013095587537dbfa4767286307fff65
10577     fixes: bz#1720566
10578     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
10579 
10580 commit 390cd44b7d0f894a05750245a87d74a7f814493d
10581 Author: Atin Mukherjee <amukherj@redhat.com>
10582 Date:   Mon Jun 24 18:32:52 2019 +0530
10583 
10584     glusterd: ignore user.* options from compatibility check in brick mux
10585     
10586     user.* options are just custom and they don't contribute anything in
10587     terms of determining the volume compatibility in brick multiplexing
10588     
10589     Fixes: bz#1723402
10590     Change-Id: Ic7e0181ab72993d29cab345cde64ae1340bf4faf
10591     Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
10592 
10593 commit a066fd70e579db9099251f1debac02f0a87de4c1
10594 Author: Amar Tumballi <amarts@redhat.com>
10595 Date:   Tue Jun 18 11:00:38 2019 +0530
10596 
10597     lcov: add more tests to glfsxmp-coverage
10598     
10599     * found a bug with quiesce fallocate() - fixed.
10600     * found a bug with cloudsync part of code in posix - fixed
10601     
10602     updates: bz#1693692
10603     Change-Id: I4f315ffebb612de072ae08761b8cd0f47714080a
10604     Signed-off-by: Amar Tumballi <amarts@redhat.com>
10605 
10606 commit fa804f13f3c1a2e8701327a3746b979f04ce09dd
10607 Author: Yaniv Kaul <ykaul@redhat.com>
10608 Date:   Thu Jun 13 16:22:44 2019 +0300
10609 
10610     glusterd.h: remove unneeded macros or move them to their users.
10611     
10612     Some macros were not used, so removed.
10613     Some macros were quite local, so moved to the respective users.
10614     Some macros simplified (removed an allocation here and there)
10615     
10616     Change-Id: Ifaf1aff15a78f105b1549ab8053378933b35df43
10617     updates: bz#1193929
10618     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
10619 
10620 commit 283b77805cca3027e333a11c9b00ac611662c9ee
10621 Author: Mohit Agrawal <moagrawal@redhat.com>
10622 Date:   Mon Jun 24 21:08:34 2019 +0530
10623 
10624     test: Fix spurious failures in bug-1040275-brick-uid-reset-on-volume-restart.t
10625     
10626     Problem: test case is failing after just starting the volume at the
10627              time of running stat command on mount point and client is
10628              getting error transport endpoint is not conencted
10629     
10630     Solution: To avoid the error make sure all brick instance should be
10631               up and mount point should be active
10632     
10633     Change-Id: I49553a04d5b13e155ee02f4a1888a07fe3ee2ff5
10634     fixes: bz#1721590
10635     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
10636 
10637 commit 24a3204d7088d11a00e4fa2145e6262f824fe0a9
10638 Author: Dan <dan@d3fy.net>
10639 Date:   Tue Jun 18 09:46:15 2019 +0530
10640 
10641     compat.h: fix a \#pragma poisoning error
10642     
10643     Fixes a failure to compile glusterfs with glibc 2.25 (on Gentoo and other systems)
10644     
10645     Making all in src
10646       CC       libglusterfs_la-dict.lo
10647     In file included from iatt.h:16:0,
10648                      from common-utils.h:44,
10649                      from dict.c:20:
10650     /usr/include/sys/sysmacros.h:57:45: error: attempt to use poisoned "system"
10651        directly.  If you did not intend to use a system-defined macro\n\
10652                                                  ^
10653     make[4]: *** [Makefile:959: libglusterfs_la-dict.lo] Error 1
10654     
10655     Fixes: bz#1494654
10656     Change-Id: I09b910b5772f52e853f87d81f3923eed9a90f7a1
10657     Signed-off-by: Amar Tumballi <amarts@redhat.com>
10658 
10659 commit d1d9c2587f5a985073039db5d15fe9adb0433547
10660 Author: Mohammed Rafi KC <rkavunga@redhat.com>
10661 Date:   Tue Jun 18 22:15:37 2019 +0530
10662 
10663     glusterd/shd: Change shd logfile to a unique name
10664     
10665     With the shd mux changes, shd was havinga a logfile
10666     with volname of the first started volume.
10667     
10668     This was creating a lot confusion, as other volumes data
10669     is also logging to a logfile which has a different vol name.
10670     
10671     With this changes the logfile will be changed to a unique name
10672     ie "/var/log/glusterfs/glustershd.log". This was the same
10673     logfile name before the shd mux
10674     
10675     Change-Id: I2b94c1f0b2cf3c9493505dddf873687755a46dda
10676     fixes: bz#1721601
10677     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
10678 
10679 commit aeb4545f35ef2740db36387cb855dcaa6bafb03d
10680 Author: Rinku Kothiya <rkothiya@redhat.com>
10681 Date:   Fri Jun 14 07:53:06 2019 +0000
10682 
10683     core: replace inet_addr with inet_pton
10684     
10685     Fixes warning raised by RPMDiff on the use of inet_addr, which may
10686     impact Ipv6 support
10687     
10688     fixes: bz#1721385
10689     
10690     Change-Id: Id2d9afa1747efa64bc79d90dd2566bff54deedeb
10691     Signed-off-by: Rinku Kothiya <rkothiya@redhat.com>
10692 
10693 commit 665d627e865d8cfa6a454c2c79eed9e9a1276865
10694 Author: Mohammed Rafi KC <rkavunga@redhat.com>
10695 Date:   Thu Jun 20 20:43:24 2019 +0530
10696 
10697     shd/mux: Fix race between mux_proc unlink and stop
10698     
10699     There is a small race window, where we have a shd proc
10700     without having a connection. That is when we stopped the
10701     last shd running on a process. The list was removed
10702     outside of a lock just after stopping the process.
10703     
10704     So there is a window where we stopped the process, but
10705     the shd proc list contains the entry.
10706     
10707     Change-Id: Id82a82509e5cd72acac24e8b7b87197626525441
10708     fixes: bz#1722541
10709     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
10710 
10711 commit e2f2f414f855f9a3da269745ae3db2d27baa7f3d
10712 Author: Pranith Kumar K <pkarampu@redhat.com>
10713 Date:   Thu Jun 20 17:05:49 2019 +0530
10714 
10715     cluster/ec: Prevent double pre-op xattrops
10716     
10717     Problem:
10718     Race:
10719     Thread-1                                    Thread-2
10720     1) Does ec_get_size_version() to perform
10721     pre-op fxattrop as part of write-1
10722                                                2) Calls ec_set_dirty_flag() in
10723                                                   ec_get_size_version() for write-2.
10724     					      This sets dirty[] to 1
10725     3) Completes executing
10726     ec_prepare_update_cbk leading to
10727     ctx->dirty[] = '1'
10728     					   4) Takes LOCK(inode->lock) to check if there are
10729     					      any flags and sets dirty-flag because
10730     				              lock->waiting_flag is 0 now. This leads to
10731     					      fxattrop to increment on-disk dirty[] to '2'
10732     
10733     At the end of the writes the file will be marked for heal even when it doesn't need heal.
10734     
10735     Fix:
10736     Perform ec_set_dirty_flag() and other checks inside LOCK() to prevent dirty[] to be marked
10737     as '1' in step 2) above
10738     
10739     Updates bz#1593224
10740     Change-Id: Icac2ab39c0b1e7e154387800fbededc561612865
10741     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
10742 
10743 commit 2959fbfbb77e2b1b2ccbc90610f8ff9362109ae3
10744 Author: Kotresh HR <khiremat@redhat.com>
10745 Date:   Thu Jun 13 16:23:21 2019 +0530
10746 
10747     posix/ctime: Fix ctime upgrade issue
10748     
10749     Problem:
10750     On a EC volume, during upgrade from the older version where
10751     ctime feature is not enabled(or not present) to the newer
10752     version where the ctime feature is available (enabled default),
10753     the self heal hangs and doesn't complete.
10754     
10755     Cause:
10756     The ctime feature has both client side code (utime) and
10757     server side code (posix). The feature is driven from client.
10758     Only if the client side sets the time in the frame, should
10759     the server side sets the time attributes in xattr. But posix
10760     setattr/fseattr was not doing that. When one of the server
10761     nodes is updated, since ctime is enabled by default, it
10762     starts setting xattr on setattr/fseattr on the updated node/brick.
10763     
10764     On a EC volume the first two updated nodes(bricks) are not a
10765     problem because there are 4 other bricks with consistent data.
10766     However once the third brick is updated, the new attribute(mdata xattr)
10767     will cause an inconsistency on metadata on 3 bricks, which
10768     prevents the file to be repaired.
10769     
10770     Fix:
10771     Don't create mdata xattr with utimes/utimensat system call.
10772     Only update if already present.
10773     
10774     Change-Id: Ieacedecb8a738bb437283ef3e0f042fd49dc4c8c
10775     fixes: bz#1720201
10776     Signed-off-by: Kotresh HR <khiremat@redhat.com>
10777 
10778 commit 804de0be6f100ede12f6131d3320cf06a227f191
10779 Author: David Spisla <david.spisla@iternity.com>
10780 Date:   Wed Jun 19 13:07:56 2019 +0200
10781 
10782     WORM-Xlator: Avoid performing fsetxattr if fd is NULL
10783     
10784     If worm_create_cbk receives an error (op_ret == -1) fd will be NULL
10785     and therefore performing fsetxattr would lead to a segfault and the
10786     brick process crashes. To avoid this we allow setting fsetxattr only
10787     if op_ret >= 0 . If an error happens we explicitly unwind
10788     
10789     Change-Id: Ie7f8a198add93e5cd908eb7029cffc834c3b58a6
10790     fixes: bz#1717757
10791     Signed-off-by: David Spisla <david.spisla@iternity.com>
10792 
10793 commit 52cf03ee314e46aa990c5b813d83019aabb04d58
10794 Author: Kinglong Mee <kinglongmee@gmail.com>
10795 Date:   Tue Apr 23 11:03:24 2019 +0800
10796 
10797     ec-heal: check file's gfid when deleting stale name
10798     
10799     A name-less lookup does not contain parent's stat,
10800     It is hard to check the lookuped file is at the right path.
10801     
10802     This patch changes to a name lookup, and check file's gfid with
10803     expected gfid. If the gfid is different, mark it estale.
10804     
10805     fixes: bz#1702131
10806     Change-Id: I2de20b10d680eed1e2fb1d3830b3b3dec4520dbf
10807     Signed-off-by: Kinglong Mee <kinglongmee@gmail.com>
10808 
10809 commit eae5f062ae008eb3b615170a102c715d70aa68e6
10810 Author: Amar Tumballi <amarts@redhat.com>
10811 Date:   Thu Jun 20 09:17:36 2019 +0530
10812 
10813     fix template file after clang-format
10814     
10815     clang-format gets applied for all the files ending with .c or .h
10816     but in this case, new-xlator.c was a template file. hence
10817     change the suffix to reflect the same, also to avoid the
10818     auto-formatting on template file.
10819     
10820     updates: bz#1193929
10821     Change-Id: I1c00a28f165f34dbe00fd3b6b070d868a56f9157
10822     Signed-off-by: Amar Tumballi <amarts@redhat.com>
10823 
10824 commit 2e11f61029d0b36893c4aa859f99c22b6202ad12
10825 Author: Mohammed Rafi KC <rkavunga@redhat.com>
10826 Date:   Wed Jun 19 15:11:59 2019 +0530
10827 
10828     afr/read: Implement latency based read child selection
10829     
10830     Network latency is an important factor selecting a read subvolume.
10831     So this patch is adding two new policy.
10832     
10833     1) We measure the latency of a child during a GF_DUMP rpc call.
10834        Then use this latency to pick a read subvol having the least
10835        latency.
10836     
10837     2) Second one is an hybrid mode where it calculates the effective
10838        latency by multiplying outstanding pending read request and
10839        latency, and choose the least one.
10840     
10841     Change-Id: Ia49c8a08ab61f7dcdad8b8950aa4d338e7accf97
10842     fixes: #520
10843     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
10844 
10845 commit bcdb77023e2efbbf06ad576851f0f38a0b8b11ab
10846 Author: Susant Palai <spalai@redhat.com>
10847 Date:   Tue Jun 18 16:43:43 2019 +0530
10848 
10849     posix: fix crash in posix_cs_set_state
10850     
10851     Fixes: bz#1721474
10852     Change-Id: Ic2a53fa3d1e9e23424c6898e0986f80d52c5e3f6
10853     Signed-off-by: Susant Palai <spalai@redhat.com>
10854 
10855 commit 1725880dabd2bac8739043c4cb5f9d844557f86e
10856 Author: Amar Tumballi <amarts@redhat.com>
10857 Date:   Mon Jun 17 16:49:48 2019 +0530
10858 
10859     encryption/crypt: remove from volume file
10860     
10861     The feature is not supported and is moved out of the codebase from
10862     glusterfs-5.x release. Doesn't make sense to keep the code to
10863     support it.
10864     
10865     For those who want to upgrade from an version supporting it to higher
10866     version, please do a 'gluster volume reset $VOL encryption reset' and
10867     then continue with the upgrade process.
10868     
10869     updates: bz#1648169
10870     Change-Id: I8cf822c0d7195940bd37f6af2432a3cac68d44d1
10871     Signed-off-by: Amar Tumballi <amarts@redhat.com>
10872 
10873 commit 6f70d4a6377365c352dad07d8eae0b2bd7af6fc6
10874 Author: Milind Changire <mchangir@redhat.com>
10875 Date:   Tue Feb 19 12:49:12 2019 +0530
10876 
10877     glusterfind: integrate with gfid2path
10878     
10879     Integration with gfid2path helps avoid file-system crawl and saves
10880     precious time. Extended attributes starting with "trusted.gfid2path."
10881     are read and the <PGFID>/<BN> values are extracted and the <PGFID> is
10882     iteratively resolved from the brick backend to arrive at the full path.
10883     
10884     Change-Id: I593b02880e3413b77bfceed4a36b00d401f03bc0
10885     fixes: #529
10886     Signed-off-by: Milind Changire <mchangir@redhat.com>
10887     Signed-off-by: Shwetha K Acharya <sacharya@redhat.com>
10888 
10889 commit 647a1b83fa39a635b01fcba38a330485fcd6afdb
10890 Author: Anoop C S <anoopcs@redhat.com>
10891 Date:   Wed Jun 12 15:41:27 2019 +0530
10892 
10893     extras/hooks: Install and package newly added post add-brick hook script
10894     
10895     https://review.gluster.org/c/glusterfs/+/22834 added a new SELinux hook
10896     script as a post add-brick operation to label new brick paths. But the
10897     change failed to install and package new script. Therefore making
10898     necessary changes to Makefile and spec file to get it installed and
10899     packaged.
10900     
10901     Change-Id: I67b8f4982c2783c34a4bc749fb4387c19a038225
10902     fixes: bz#1717953
10903     Signed-off-by: Anoop C S <anoopcs@redhat.com>
10904 
10905 commit b99f6eba2e4d53ba8220034d9b8a4530681ec77e
10906 Author: Kinglong Mee <mijinlong@open-fs.com>
10907 Date:   Tue Feb 26 15:20:33 2019 +0800
10908 
10909     md-cache: only update generation for inode at upcall and NULL stat
10910     
10911     1. For parallel writes from nfs-ganesha, two fops with two generations,
10912        but the fops reply maybe returned disordered.
10913     2. The inode md-cache timeout should not increase conf->generation.
10914     
10915     With this patch,
10916     1, Fop only gets generation from inode md-cache or conf, does not increase it.
10917     2. The generation is increased at upcall invalidate, estal/enoent error
10918        invalidate, reply with zeroed out stat from write-behind.
10919     
10920     Change-Id: I897ecaa143fd18bc024c1948c7d1a6f831fd53da
10921     Updates: bz#1683594
10922     Signed-off-by: Kinglong Mee <mijinlong@open-fs.com>
10923 
10924 commit c9097077922647b61570f087d16103f1c7aced99
10925 Author: N Balachandran <nbalacha@redhat.com>
10926 Date:   Tue Jun 18 15:33:29 2019 +0530
10927 
10928     cluster/dht: Strip out dht xattrs
10929     
10930     Some internal DHT xattrs were not being
10931     removed when calling getxattr in pass-through mode.
10932     This has been fixed.
10933     
10934     Change-Id: If7e3dbc7b495db88a566bd560888e3e9c167defa
10935     fixes: bz#1721435
10936     Signed-off-by: N Balachandran <nbalacha@redhat.com>
10937 
10938 commit c8f302b7b34b45b2cf340634ed66de899f900c23
10939 Author: Sunny Kumar <sunkumar@redhat.com>
10940 Date:   Tue Jun 18 16:25:35 2019 +0530
10941 
10942     geo-rep: Fix permissions for GEOREP_DIR in non-root setup
10943     
10944     During mountbroker setup: 'gluster-mountbroker <mountbroker-root> <group>'
10945     commad to set the permission and group for GEOREP_DIR directory
10946     (/var/lib/glusterd/geo-replication) fails due to extra argument, which is
10947     enssential for non-root geo-rep setup.
10948     
10949     fixes: bz#1721441
10950     Change-Id: Ia83442733bf0b29f630e8c9e398097316efca092
10951     Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
10952 
10953 commit 2d278f0407ab7d29507dc697653b39d72ddee472
10954 Author: Yaniv Kaul <ykaul@redhat.com>
10955 Date:   Sun May 26 11:18:05 2019 +0300
10956 
10957     glusterd-volgen.c: remove BD xlator from the graph
10958     
10959     The BD xlator was removed some time ago. Remove it from the graph.
10960     We can also remove the caps settings - only the BD xlator
10961     was using it.
10962     
10963     Lastly, remove the caps (which only BD was using) and the document
10964     describing the translator.
10965     
10966     Change-Id: Id0adcb2952f4832a5dc6301e726874522e07935d
10967     updates: bz#1193929
10968     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
10969 
10970 commit b79222f650f6613758f036c749f5f879a54148a7
10971 Author: SheetalPamecha <spamecha@redhat.com>
10972 Date:   Mon Jun 17 15:42:20 2019 +0530
10973 
10974     core: fedora 30 compiler warnings
10975     
10976     warning: ‘%s’ directive argument is null [-Wformat-overflow=]
10977     
10978     Change-Id: I69b8d47f0002c58b00d1cc947fac6f1c64e0b295
10979     updates: bz#1193929
10980     Signed-off-by: SheetalPamecha <spamecha@redhat.com>
10981 
10982 commit 25ad5aca23b257cdd129cd1d4518b048fbba87bb
10983 Author: Mohit Agrawal <moagrawal@redhat.com>
10984 Date:   Mon Jun 17 11:10:42 2019 +0530
10985 
10986     tests: subdir-mount.t is failing for brick_mux regrssion
10987     
10988     To avoid the failure wait to run hook script S13create-subdir-mounts.sh
10989     after executed add-brick command by test case.
10990     
10991     Change-Id: I063b6d0f86a550ed0a0527255e4dfbe8f0a8c02e
10992     fixes: bz#1720993
10993     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
10994 
10995 commit 7cc20c1eb09c041b331c4add6c24212fbdcda3b4
10996 Author: Xavier Hernandez <jahernan@redhat.com>
10997 Date:   Fri Jan 19 12:18:13 2018 +0100
10998 
10999     core: improve timer accuracy
11000     
11001     Also fixed some issues on test ec-1468261.t.
11002     
11003     Change-Id: If156f86af986d9eed13cdd1f15c5a7214cd11706
11004     Updates: bz#1193929
11005     Signed-off-by: Xavier Hernandez <jahernan@redhat.com>
11006 
11007 commit 513df02b19af14b8b006c5c2e60c2a7447146aa2
11008 Author: Sanju Rakonde <srakonde@redhat.com>
11009 Date:   Thu Jun 13 16:28:04 2019 +0530
11010 
11011     glusterd: log error message only when rsp.op_ret is negative
11012     
11013     Problem: commit d42221bec9 added a log message based on rsp.op_ret
11014     check. but while running subdir-mount.t, this message is seen even on
11015     successful mounts.
11016     
11017     Solution: in __server_getspec(), return value of sys_read() is assigned to
11018     ret, which will be a non-negative number in when sys_read() is success.
11019     This non-zero value is assigned to rsp.op_ret. We should log an error only
11020     when rsp.op_ret is negative.
11021     
11022     fixes: bz#1718848
11023     
11024     Change-Id: Ieef8ba33c2c7b4a97d4aef17543f58e66fd3b341
11025     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
11026 
11027 commit 1984fddf10fd3ed8ddedfab45424db9656b271a8
11028 Author: Atin Mukherjee <amukherj@redhat.com>
11029 Date:   Tue Jun 11 09:52:06 2019 +0530
11030 
11031     glusterd: add GF_TRANSPORT_BOTH_TCP_RDMA in glusterd_get_gfproxy_client_volfile
11032     
11033     ... with out which volume creation fails with "volume create: <xyz>: failed:
11034     Failed to create volume files"
11035     
11036     Fixes: bz#1716812
11037     Change-Id: I2f4c2c6d5290f066b54e1c1db19e25db9937bedb
11038     Signed-off-by: Atin Mukherjee <amukherj@redhat.com>
11039 
11040 commit 01846e7b9b0b92e4a10f5cbcc921bad90441660d
11041 Author: Kotresh HR <khiremat@redhat.com>
11042 Date:   Thu Jun 13 22:43:47 2019 +0530
11043 
11044     uss: Fix tar issue with ctime and uss enabled
11045     
11046     Problem:
11047     If ctime and uss enabled, tar still complains with 'file
11048     changed as we read it'
11049     
11050     Cause:
11051     To clear nfs cache (gluster-nfs), the ctime was incremented
11052     in snap-view client on stat cbk.
11053     
11054     Fix:
11055     The ctime should not be incremented manually. Since gluster-nfs
11056     is planning to be deprecated, this code is being removed to
11057     fix the issue.
11058     
11059     Change-Id: Iae7f100c20fce880a50b008ba716077350281404
11060     fixes: bz#1720290
11061     Signed-off-by: Kotresh HR <khiremat@redhat.com>
11062 
11063 commit b460e58d07fac340b713778f979d23705a8b70ff
11064 Author: Mohammed Rafi KC <rkavunga@redhat.com>
11065 Date:   Tue Jun 4 05:37:57 2019 +0530
11066 
11067     afr/fini: Free local_pool data during an afr fini
11068     
11069     We should free the mem_pool local_pool during an afr_fini.
11070     Otherwise this will lead to mem leak for shd
11071     
11072     Change-Id: I805a34a88077bf7b886c28b403798bf9eeeb1c0b
11073     Updates: bz#1716695
11074     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
11075 
11076 commit 632c10b5e976079d825f8808b5e33ab2e1dda469
11077 Author: Amar Tumballi <amarts@redhat.com>
11078 Date:   Fri Jun 14 15:00:12 2019 +0530
11079 
11080     cli: don't fail if logging initialize fails
11081     
11082     in many cases, gluster's cli can run as non-privileged mode
11083     (like in geo-rep non-root setup). Just because logging fails
11084     in cli, lets not fail the overall process. Not much of
11085     debugging help in CLI logs anyways. Most of the debugging
11086     happens once the call reaches server (glusterd).
11087     
11088     Fixes: bz#1535511
11089     Change-Id: I9f07c61b8c3acc95ec08230ff539a35dfd0ff9dc
11090     Signed-off-by: Amar Tumballi <amarts@redhat.com>
11091 
11092 commit 1cb32e3df2775dabc57df6931ffce96aece7109e
11093 Author: Sheetal Pamecha <spamecha@redhat.com>
11094 Date:   Fri May 31 17:21:22 2019 +0530
11095 
11096     geo-rep/gsyncd: name is not freed in one of the cases
11097     
11098     CID: 1400730
11099     updates: bz#789278
11100     Change-Id: I0f6924050a31d3d2cc0b555f859920e349728e0a
11101     Signed-off-by: Sheetal Pamecha <spamecha@redhat.com>
11102 
11103 commit 82b8458ac6d31c72dc7a2f69df7f50bb1e4b7618
11104 Author: Niels de Vos <ndevos@redhat.com>
11105 Date:   Fri Jun 14 12:43:45 2019 +0200
11106 
11107     build: always build glusterfs-cli to allow monitoring/managing from clients
11108     
11109     Fixes: bz#1720615
11110     Change-Id: I5071f3255ff615113b36b08cd5326be6e37d907d
11111     Signed-off-by: Niels de Vos <ndevos@redhat.com>
11112 
11113 commit e7211901e9f987d0efcd7d699c995634fc9f24d6
11114 Author: Aravinda VK <avishwan@redhat.com>
11115 Date:   Fri Jun 14 17:34:34 2019 +0530
11116 
11117     tests: Add missing NFS test tag to the testfile
11118     
11119     $SRC/glusterfs/bugs/nfs/showmount-many-clients.t
11120     
11121     Change-Id: I48758cc66fcb55f48c4a8a0a738b06867f6814a1
11122     Signed-off-by: Aravinda VK <avishwan@redhat.com>
11123     Updates: bz#1193929
11124 
11125 commit 06ee9e13db03a02b343a5d44cae3c4a5da48bf8a
11126 Author: Amar Tumballi <amarts@redhat.com>
11127 Date:   Fri Jun 14 12:53:15 2019 +0530
11128 
11129     clang-scan: resolve warning
11130     
11131     dht-common.c: because there was a 'goto err' before assigning
11132                   the 'local' variable, there is possibility of NULL
11133                   dereference. As the check which was done wouldn't
11134                   ever be true, removed the check.
11135     
11136     glusterd-geo-rep.c: a possible path where 'slave_host' could be
11137                  NULL when it gets passed to strcmp() is found.
11138                  strcmp() expects a valid string. Add a NULL check.
11139     
11140     Updates: bz#1622665
11141     Change-Id: I64c280bc1beac9a2b109e8fa88f2a5ce8b823c3a
11142     Signed-off-by: Amar Tumballi <amarts@redhat.com>
11143 
11144 commit 836c7ec096ebff089482f1e5a25fa55df28b14ea
11145 Author: Amar Tumballi <amarts@redhat.com>
11146 Date:   Fri Jun 14 22:38:10 2019 +0530
11147 
11148     glfs: add syscall.h after header cleanup
11149     
11150     in one of the recent patches, we cleaned-up the unneccesary header
11151     file includes. In the order of merging the patches, there cropped
11152     up an compile error.
11153     
11154     updates: bz#1193929
11155     Change-Id: I2ad52aa918f9c698d5273bb293838de6dd50ac31
11156     Signed-off-by: Amar Tumballi <amarts@redhat.com>
11157 
11158 commit 0a6fe8551ac9807a8b6ad62241ec8048cf9f9025
11159 Author: Yaniv Kaul <ykaul@redhat.com>
11160 Date:   Sun Jun 9 13:31:31 2019 +0300
11161 
11162     multiple files: another attempt to remove includes
11163     
11164     There are many include statements that are not needed.
11165     A previous more ambitious attempt failed because of *BSD plafrom
11166     (see https://review.gluster.org/#/c/glusterfs/+/21929/ )
11167     
11168     Now trying a more conservative reduction.
11169     It does not solve all circular deps that we have, but it
11170     does reduce some of them. There is just too much to handle
11171     reasonably (dht-common.h includes dht-lock.h which includes
11172     dht-common.h ...), but it does reduce the overall number of lines
11173     of include we need to look at in the future to understand and fix
11174     the mess later one.
11175     
11176     Change-Id: I550cd001bdefb8be0fe67632f783c0ef6bee3f9f
11177     updates: bz#1193929
11178     Signed-off-by: Yaniv Kaul <ykaul@redhat.com>
11179 
11180 commit 65e3624091e895ca32cd91468552589f34d17595
11181 Author: Aravinda VK <avishwan@redhat.com>
11182 Date:   Fri Jun 14 16:11:37 2019 +0530
11183 
11184     eventsapi: Fix Exception class for Python3
11185     
11186     Python2 exception provides message attribute for custom exceptions.
11187     But it is not available in Python3. Add init method for custom exception
11188     to handle the same.
11189     
11190     Original crash(IndexError reported in the bug) was fixed with
11191     https://review.gluster.org/#/c/glusterfs/+/22294/ But that patch only
11192     works in Python2 and fails in Python3 since Exception in Python 3
11193     doesn't have "message" attribute.
11194     
11195     Fixes: bz#1573226
11196     Change-Id: If9117048f9ff0615f5da1880075ec12c0ff4855e
11197     Signed-off-by: Aravinda VK <avishwan@redhat.com>
11198 
11199 commit 6eaa561526c340c48b12f0cf149c6c5a2903d751
11200 Author: Amar Tumballi <amarts@redhat.com>
11201 Date:   Thu Mar 14 10:04:28 2019 +0530
11202 
11203     gfapi: provide an api for setting statedump path
11204     
11205     Currently for an application using glfsapi to use glusterfs, when a
11206     statedump is taken, it uses /var/run/gluster dir to dump info.
11207     
11208     There can be concerns as this directory may be owned by some other
11209     user, and hence it may fail taking statedump. Such applications
11210     should have an option to use different path.
11211     
11212     This patch provides an API to do so.
11213     
11214     Updates: bz#1689097
11215     Change-Id: I8918e002bc823d83614c972b6c738baa04681b23
11216     Signed-off-by: Amar Tumballi <amarts@redhat.com>
11217 
11218 commit 928395eddf4241433effc55e294a7996108da7d8
11219 Author: Xie Changlong <xiechanglong@cmss.chinamobile.com>
11220 Date:   Mon Jun 3 15:59:38 2019 +0800
11221 
11222     gnfs: support inode dump
11223     
11224     So, we will get more debug info.
11225     
11226     fixes: #679
11227     Change-Id: I3588e204ad25c20b69271c1a4ee17d0d158bd794
11228     Signed-off-by: Xie Changlong <xiechanglong@cmss.chinamobile.com>
11229 
11230 commit 675dd98ce0c772b732c7f45a8e51dcf0d64f57fc
11231 Author: Soumya Koduri <skoduri@redhat.com>
11232 Date:   Fri Jun 7 19:33:07 2019 +0530
11233 
11234     upcall: Avoid sending notifications for invalid inodes
11235     
11236     For nameless LOOKUPs, server creates a new inode which shall
11237     remain invalid until the fop is successfully processed post
11238     which it is linked to the inode table.
11239     
11240     But incase if there is an already linked inode for that entry,
11241     it discards that newly created inode which results in upcall
11242     notification. This may result in client being bombarded with
11243     unnecessary upcalls affecting performance if the data set is huge.
11244     
11245     This issue can be avoided by looking up and storing the upcall
11246     context in the original linked inode (if exists), thus saving up on
11247     those extra callbacks.
11248     
11249     Change-Id: I044a1737819bb40d1a049d2f53c0566e746d2a17
11250     fixes: bz#1718338
11251     Signed-off-by: Soumya Koduri <skoduri@redhat.com>
11252 
11253 commit bc6fd4cfa6ed34de3ffc02e2279fcc713f80f530
11254 Author: Soumya Koduri <skoduri@redhat.com>
11255 Date:   Fri Jun 7 17:20:15 2019 +0530
11256 
11257     gfapi: fix incorrect initialization of upcall syncop arguments
11258     
11259     While sending upcall notifications via synctasks, the argument used to
11260     carry relevant data for these tasks is not initialized properly. This patch
11261     is to fix the same.
11262     
11263     Change-Id: I9fa8f841e71d3c37d3819fbd430382928c07176c
11264     fixes: bz#1718316
11265     Signed-off-by: Soumya Koduri <skoduri@redhat.com>
11266 
11267 commit 4d51b878c9c41e81f965569a4fe19a95a0a42c25
11268 Author: Shwetha K Acharya <sacharya@redhat.com>
11269 Date:   Mon Jun 3 18:05:24 2019 +0530
11270 
11271     cli: Remove-brick warning seems unnecessary
11272     
11273     As force-migration option is disabled by default,
11274     the warning seems unnessary.
11275     
11276     Rephrased the warning to make best sense out of it.
11277     
11278     fixes: bz#1712668
11279     Change-Id: Ia18c3c5e7b3fec808fce2194ca0504a837708822
11280     Signed-off-by: Shwetha K Acharya <sacharya@redhat.com>
11281 
11282 commit 582f91fcad42b6b7da653d2a587d196a7a1e0204
11283 Author: Anoop C S <anoopcs@redhat.com>
11284 Date:   Thu Jun 6 18:33:19 2019 +0530
11285 
11286     extras/hooks: Add SELinux label on new bricks during add-brick
11287     
11288     Change-Id: Ifd8ae5eeb91b968cc1a9a9b5d15844c5233d56db
11289     fixes: bz#1717953
11290     Signed-off-by: Anoop C S <anoopcs@redhat.com>
11291 
11292 commit 3dd03146bb7037ae2ebea0579d0b81be27fdd927
11293 Author: Sunny Kumar <sunkumar@redhat.com>
11294 Date:   Mon May 13 14:32:53 2019 +0530
11295 
11296     geo-rep : fix mountbroker setup
11297     
11298     Problem:
11299     
11300     Unable to setup mountbroker root directory while creating geo-replication
11301     session for non-root user.
11302     
11303     Casue:
11304     With patch[1] which defines the max-port for glusterd one extra sapce
11305     got added in field of 'option max-port'.
11306     [1]. https://review.gluster.org/#/c/glusterfs/+/21872/
11307     
11308     In geo-rep spliting of key-value pair form vol file was done on the
11309     basis of space so this additional space caused "ValueError: too many values
11310     to unpack".
11311     
11312     Solution:
11313     Use split so that it can treat consecutive whitespace as a single separator.
11314     
11315     Fixes: bz#1709248
11316     Change-Id: Ia22070a43f95d66d84cb35487f23f9ee58b68c73
11317     Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
11318 
11319 commit 952cf7e4f4393fcd9cf8c16b013d8f28915c990e
11320 Author: Xavi Hernandez <xhernandez@redhat.com>
11321 Date:   Fri May 31 18:40:30 2019 +0200
11322 
11323     libglusterfs: cleanup iovec functions
11324     
11325     This patch cleans some iovec code and creates two additional helper
11326     functions to simplify management of iovec structures.
11327     
11328       iov_range_copy(struct iovec *dst, uint32_t dst_count, uint32_t dst_offset,
11329                      struct iovec *src, uint32_t src_count, uint32_t src_offset,
11330                      uint32_t size);
11331     
11332         This function copies up to 'size' bytes from 'src' at offset
11333         'src_offset' to 'dst' at 'dst_offset'. It returns the number of
11334         bytes copied.
11335     
11336       iov_skip(struct iovec *iovec, uint32_t count, uint32_t size);
11337     
11338         This function removes the initial 'size' bytes from 'iovec' and
11339         returns the updated number of iovec vectors remaining.
11340     
11341     The signature of iov_subset() has also been modified to make it safer
11342     and easier to use. The new signature is:
11343     
11344       iov_subset(struct iovec *src, int src_count, uint32_t start, uint32_t size,
11345                  struct iovec **dst, int32_t dst_count);
11346     
11347         This function creates a new iovec array containing the subset of the
11348         'src' vector starting at 'start' with size 'size'. The resulting
11349         array is allocated if '*dst' is NULL, or copied to '*dst' if it fits
11350         (based on 'dst_count'). It returns the number of iovec vectors used.
11351     
11352     A new set of functions to iterate through an iovec array have been
11353     created. They can be used to simplify the implementation of other
11354     iovec-based helper functions.
11355     
11356     Change-Id: Ia5fe57e388e23392a8d6cdab17670e337cadd587
11357     Updates: bz#1193929
11358     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
11359 
11360 commit c525fdc6bd79684da196e0fe7ca5630c1bac8f21
11361 Author: Amar Tumballi <amarts@redhat.com>
11362 Date:   Mon Jun 10 09:39:38 2019 +0530
11363 
11364     tests: keep glfsxmp in tests directory
11365     
11366     this is critical so all the tests will be contained in the same
11367     directory, and one can just 'cp -a tests/ <any-location>/' and
11368     run glusterfs tests.
11369     
11370     only 'glfsxmp.c' was an exception as it was just copying the
11371     file from api example directory. Now moved it to tests.
11372     
11373     updates: bz#1193929
11374     Change-Id: I00359d64be580bffc5b3c3a090968d86c2c6952a
11375     Signed-off-by: Amar Tumballi <amarts@redhat.com>
11376 
11377 commit 8e5656d950ff7bca21594b47921c6c0bf98eecea
11378 Author: karthik-us <ksubrahm@redhat.com>
11379 Date:   Mon Jun 10 23:19:44 2019 +0530
11380 
11381     tests: Fix split-brain-favorite-child-policy.t failure
11382     
11383     Problem:
11384     The test case is failing to heal the volume within $HEAL_TIMEOUT @195.
11385     This is happening because as part of split-brain resolution the file
11386     gets expunged from the sink and the new entry mark for that file will
11387     be done on the source bricks as part of impunging. Since the source
11388     bricks shd-threads failed to get the heal-domain lock, they will wait
11389     for the heal-timeout of 10 minutes, which is greater than $HEAL_TIMEOUT.
11390     
11391     Fix:
11392     Set the cluster.heal-timeout to 5 seconds to trigger the heal so that
11393     one of the source brick heals the file within the $HEAL_TIMEOUT.
11394     
11395     Change-Id: Ie73c578cc5361c0d617a48ccc86026734d20ba8c
11396     fixes: bz#1718998
11397     Signed-off-by: karthik-us <ksubrahm@redhat.com>
11398 
11399 commit 1b0b869d91d4e5bedc69922128551602dc4bbc13
11400 Author: karthik-us <ksubrahm@redhat.com>
11401 Date:   Thu Jun 6 10:59:42 2019 +0530
11402 
11403     Cluster/afr: Don't treat all bricks having metadata pending as split-brain
11404     
11405     Problem:
11406     We currently don't have a roll-back/undoing of post-ops if quorum is not met.
11407     Though the FOP is still unwound with failure, the xattrs remain on the disk.
11408     Due to these partial post-ops and partial heals (healing only when 2 bricks
11409     are up), we can end up in metadata split-brain purely from the afr xattrs
11410     point of view i.e each brick is blamed by atleast one of the others for
11411     metadata. These scenarios are hit when there is frequent connect/disconnect
11412     of the client/shd to the bricks.
11413     
11414     Fix:
11415     Pick a source based on the xattr values. If 2 bricks blame one, the blamed
11416     one must be treated as sink. If there is no majority, all are sources. Once
11417     we pick a source, self-heal will then do the heal instead of erroring out
11418     due to split-brain.
11419     This patch also adds restriction of all the bricks to be up to perform
11420     metadata heal to avoid any metadata loss.
11421     
11422     Removed the test case tests/bugs/replicate/bug-1468279-source-not-blaming-sinks.t
11423     as it was doing metadata heal even when only 2 of 3 bricks were up.
11424     
11425     Change-Id: I07a9d62f84ceda329dcab1f02a33aeed258dcb09
11426     fixes: bz#1717819
11427     Signed-off-by: karthik-us <ksubrahm@redhat.com>
11428 
11429 commit 5b35b18be6a04ed4dd69dc71e5fb436c212c15fc
11430 Author: Sunny Kumar <sunkumar@redhat.com>
11431 Date:   Mon Jun 10 15:16:48 2019 +0530
11432 
11433     tests: added cleanup for lock files
11434     
11435     Problem: useradd fails with: Cannot allocate memory
11436              useradd: cannot lock /etc/passwd; try again later.
11437     
11438     Solution:
11439     
11440     Lock files should get automatically removed once "usradd" or "groupadd"
11441     command finishes. But sometimes we encounter situations (bugs) where
11442     some of these files may not get properly unlocked after the execution of
11443     the command. In that case, when we execute useradd next time, it may show
11444     the error “cannot lock /etc/password” or “unable to lock group file”.
11445     So, to avoid any such errors, check for any lock files under /etc and
11446     remove those.
11447     
11448     updates: bz#1193929
11449     
11450     Change-Id: If6456a271c2bc0717f768d7101a40ce44a9af3d7
11451     Signed-off-by: Sunny Kumar <sunkumar@redhat.com>
11452 
11453 commit dc119e9c2f3db6d029ab1c5a81c171180db58192
11454 Author: Krutika Dhananjay <kdhananj@redhat.com>
11455 Date:   Fri Apr 5 12:29:23 2019 +0530
11456 
11457     features/shard: Fix extra unref when inode object is lru'd out and added back
11458     
11459     Long tale of double unref! But do read...
11460     
11461     In cases where a shard base inode is evicted from lru list while still
11462     being part of fsync list but added back soon before its unlink, there
11463     could be an extra inode_unref() leading to premature inode destruction
11464     leading to crash.
11465     
11466     One such specific case is the following -
11467     
11468     Consider features.shard-deletion-rate = features.shard-lru-limit = 2.
11469     This is an oversimplified example but explains the problem clearly.
11470     
11471     First, a file is FALLOCATE'd to a size so that number of shards under
11472     /.shard = 3 > lru-limit.
11473     Shards 1, 2 and 3 need to be resolved. 1 and 2 are resolved first.
11474     Resultant lru list:
11475                                    1 -----> 2
11476     refs on base inode -          (1)  +   (1) = 2
11477     3 needs to be resolved. So 1 is lru'd out. Resultant lru list -
11478     		               2 -----> 3
11479     refs on base inode -          (1)  +   (1) = 2
11480     
11481     Note that 1 is inode_unlink()d but not destroyed because there are
11482     non-zero refs on it since it is still participating in this ongoing
11483     FALLOCATE operation.
11484     
11485     FALLOCATE is sent on all participant shards. In the cbk, all of them are
11486     added to fync_list.
11487     Resulting fsync list -
11488                                    1 -----> 2 -----> 3 (order doesn't matter)
11489     refs on base inode -          (1)  +   (1)  +   (1) = 3
11490     Total refs = 3 + 2 = 5
11491     
11492     Now an attempt is made to unlink this file. Background deletion is triggered.
11493     The first $shard-deletion-rate shards need to be unlinked in the first batch.
11494     So shards 1 and 2 need to be resolved. inode_resolve fails on 1 but succeeds
11495     on 2 and so it's moved to tail of list.
11496     lru list now -
11497                                   3 -----> 2
11498     No change in refs.
11499     
11500     shard 1 is looked up. In lookup_cbk, it's linked and added back to lru list
11501     at the cost of evicting shard 3.
11502     lru list now -
11503                                   2 -----> 1
11504     refs on base inode:          (1)  +   (1) = 2
11505     fsync list now -
11506                                   1 -----> 2 (again order doesn't matter)
11507     refs on base inode -         (1)  +   (1) = 2
11508     Total refs = 2 + 2 = 4
11509     After eviction, it is found 3 needs fsync. So fsync is wound, yet to be ack'd.
11510     So it is still inode_link()d.
11511     
11512     Now deletion of shards 1 and 2 completes. lru list is empty. Base inode unref'd and
11513     destroyed.
11514     In the next batched deletion, 3 needs to be deleted. It is inode_resolve()able.
11515     It is added back to lru list but base inode passed to __shard_update_shards_inode_list()
11516     is NULL since the inode is destroyed. But its ctx->inode still contains base inode ptr
11517     from first addition to lru list for no additional ref on it.
11518     lru list now -
11519                                   3
11520     refs on base inode -         (0)
11521     Total refs on base inode = 0
11522     Unlink is sent on 3. It completes. Now since the ctx contains ptr to base_inode and the
11523     shard is part of lru list, base shard is unref'd leading to a crash.
11524     
11525     FIX:
11526     When shard is readded back to lru list, copy the base inode pointer as is into its inode ctx,
11527     even if it is NULL. This is needed to prevent double unrefs at the time of deleting it.
11528     
11529     Change-Id: I99a44039da2e10a1aad183e84f644d63ca552462
11530     Updates: bz#1696136
11531     Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
11532 
11533 commit 4cfc5788af2488d173ac033850370c4f9ed7a05e
11534 Author: Mohammed Rafi KC <rkavunga@redhat.com>
11535 Date:   Sun Jun 2 01:36:33 2019 +0530
11536 
11537     ec/fini: Fix race between xlator cleanup and on going async fop
11538     
11539     Problem:
11540     While we process a cleanup, there is a chance for a race between
11541     async operations, for example ec_launch_replace_heal. So this can
11542     lead to invalid mem access.
11543     
11544     Solution:
11545     Just like we track on going heal fops, we can also track fops like
11546     ec_launch_replace_heal, so that we can decide when to send a
11547     PARENT_DOWN request.
11548     
11549     Change-Id: I055391c5c6c34d58aef7336847f3b570cb831298
11550     fixes: bz#1703948
11551     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
11552 
11553 commit 9e0de2b634b888dd069e908b7745197d20fe7036
11554 Author: Mohammed Rafi KC <rkavunga@redhat.com>
11555 Date:   Tue Jun 4 11:13:50 2019 +0530
11556 
11557     xlator/log: Add more logging in xlator_is_cleanup_starting
11558     
11559     This patch will add two extra logs for invalid argument
11560     
11561     Change-Id: I3950b4f4b9d88b1f1e788ef93d8f09d4bd8d4d8b
11562     updates: bz#1703948
11563     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
11564 
11565 commit 2eaf8e846afd71c30f2a6ff6f863a39b1145b8b6
11566 Author: kshithijiyer <kshithij.ki@gmail.com>
11567 Date:   Wed Jun 5 19:40:29 2019 +0530
11568 
11569     Fixing formatting errors in markdown files
11570     
11571     There are a lot of fromatting error is markdown
11572     files peresent under /doc directiory of the project.
11573     Fixing formatting errors and sending a patch.
11574     
11575     Fixes: bz#1718273
11576     Change-Id: I08f938088bbaaafddf634f73616ea0dbfe7aedf3
11577     Signed-off-by: kshithijiyer <kshithij.ki@gmail.com>
11578 
11579 commit 2ff76fa45c53b8e291cf70c98b28800f3ed5f6fc
11580 Author: Raghavendra Bhat <raghavendra@redhat.com>
11581 Date:   Tue May 14 10:56:15 2019 -0400
11582 
11583     uss: Ensure that snapshot is deleted before creating a new snapshot
11584     
11585     * Also some logging enhancements in snapview-server
11586     
11587     Change-Id: I6a7646771cedf4bd1c62806eea69d720bbaf0c83
11588     fixes: bz#1715921
11589     Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
11590 
11591 commit fa32b594d12673698da0c1dab8491db8655b4d50
11592 Author: Ashish Pandey <aspandey@redhat.com>
11593 Date:   Tue Jun 4 11:17:21 2019 +0530
11594 
11595     cluster/replicate: Modify command in unit file to assign port correctly
11596     
11597     Problem:
11598     In unit file of TA process we have been using ta-vol as
11599     volume id and also ta-vol-server.transport.socket.listen-port=24007
11600     
11601     In volume file for TA process we only consider volname
11602     as "ta" and not as "ta-vol". That's why it was not able to assign
11603     this port number to ta process as in volume file it will try to
11604     find server xlator as ta-vol
11605     
11606     volume ta-server  <<<<<<<<< not ta-vol-server
11607      46     type protocol/server
11608      47     option transport.listen-backlog 10
11609      48     option transport.socket.keepalive-count 9
11610      49     option transport.socket.keepalive-interval 2
11611      50     option transport.socket.keepalive-time 20
11612      51     option transport.tcp-user-timeout 0
11613      52     option transport.socket.keepalive 1
11614      53     option auth.addr./mnt/thin-arbiter.allow *
11615      54     option auth-path /mnt/thin-arbiter
11616      55     option transport.address-family inet
11617      56     option transport-type tcp
11618      57     subvolumes ta-io-stats
11619      58 end-volume
11620     
11621     Solution:
11622     Provide "ta" as vol id for the  command which Unit file
11623     is going to execute.
11624     
11625     Also, made changes in setup-thin-arbiter.sh to correctly
11626     identify the directory of Unit file irrespective of the location from
11627     where we are executing this script.
11628     
11629     Change-Id: Ia7bbccdc0304e7dfaaa732bebb726fba731d1d33
11630     fixes: bz#1716766
11631     Signed-off-by: Ashish Pandey <aspandey@redhat.com>
11632 
11633 commit ecafe78378209ea28201624fca262bfdbd425365
11634 Author: Ravishankar N <ravishankar@redhat.com>
11635 Date:   Thu Jun 6 13:19:29 2019 +0530
11636 
11637     glusterd: store fips-mode-rchecksum option in the info file
11638     
11639     commit 146e4b45d0ce906ae50fd6941a1efafd133897ea enabled
11640     storage.fips-mode-rchecksum option for all new volumes with op-version
11641     >=GD_OP_VERSION_7_0 but `gluster vol get $volname
11642     storage.fips-mode-rchecksum` was displaying it as 'off'. This patch fixes it.
11643     
11644     fixes: bz#1717782
11645     Change-Id: Ie09f89838893c5776a3f60569dfe8d409d1494dd
11646     Signed-off-by: Ravishankar N <ravishankar@redhat.com>
11647 
11648 commit fdc596731ea49af87a1dfdc4f255459e5fda60aa
11649 Author: Anoop C S <anoopcs@redhat.com>
11650 Date:   Thu Jun 6 15:46:15 2019 +0530
11651 
11652     glusterd: Fix typos
11653     
11654     Change-Id: I8cf0a153f84ef2d162e6dd03261441d211c07d40
11655     updates: bz#1193929
11656     Signed-off-by: Anoop C S <anoopcs@redhat.com>
11657 
11658 commit 89a34b574ee86b77fbae5295c766471086b9ea8f
11659 Author: Kotresh HR <khiremat@redhat.com>
11660 Date:   Thu Jun 6 12:54:04 2019 +0530
11661 
11662     tests/utils: Fix py2/py3 util python scripts
11663     
11664     Following files are fixed.
11665     
11666     tests/bugs/distribute/overlap.py
11667     tests/utils/changelogparser.py
11668     tests/utils/create-files.py
11669     tests/utils/gfid-access.py
11670     tests/utils/libcxattr.py
11671     
11672     Change-Id: I3db857cc19e19163d368d913eaec1269fbc37140
11673     updates: bz#1193929
11674     Signed-off-by: Kotresh HR <khiremat@redhat.com>
11675 
11676 commit a981e7cc12bb89304e8736d857f23f3a394512d0
11677 Author: Amar Tumballi <amarts@redhat.com>
11678 Date:   Fri Jun 7 13:55:12 2019 +0530
11679 
11680     tests/quick-read-upcall: mark it bad
11681     
11682     Frequent intermittent failures observed.
11683     ```
11684     08:59:24 ok  11 [     10/      3] <  36> 'write_to /mnt/glusterfs/0/test.txt test-message1'
11685     08:59:24 ok  12 [     10/      6] <  37> 'test-message1 cat /mnt/glusterfs/0/test.txt'
11686     08:59:24 ok  13 [     10/      4] <  38> 'test-message0 cat /mnt/glusterfs/1/test.txt'
11687     08:59:24 not ok  14 [   3715/      6] <  45> 'test-message1 cat /mnt/glusterfs/1/test.txt' -> 'Got "test-message0" instead of "test-message1"'
11688     08:59:24 ok  15 [     10/    162] <  47> 'gluster --mode=script --wignore volume set patchy features.cache-invalidation on'
11689     08:59:24 ok  16 [     10/    148] <  48> 'gluster --mode=script --wignore volume set patchy performance.qr-cache-timeout 15'
11690     ```
11691     
11692     updates: bz#1718191
11693     Change-Id: Ieb9e5a9a428995ff178f77bc4a5155b8298d3fa0
11694     Signed-off-by: Amar Tumballi <amarts@redhat.com>
11695 
11696 commit edc2910a8ab3a7d9cfbb19e3f20179c0799c0c1b
11697 Author: Amar Tumballi <amarts@redhat.com>
11698 Date:   Fri Jun 7 13:58:48 2019 +0530
11699 
11700     tests/volume-scale-shd-mux: mark as bad test
11701     
11702     The test is giving frequent failures in regression.
11703     
11704     Error seen is normally like below:
11705     
11706     `09:09:24 not ok  58 [     14/  80343] < 104> '^3$ number_healer_threads_shd patchy_distribute1 __afr_shd_healer_wait' -> 'Got "1" instead of "^3$"'`
11707     
11708     updates: bz#1708929
11709     Change-Id: I240bdcfb76b1f953d75937a53c5dfabba134f282
11710     Signed-off-by: Amar Tumballi <amarts@redhat.com>
11711 
11712 commit 373f045497b3e696ab6492b9c8ea105ffe947b91
11713 Author: Mohammed Rafi KC <rkavunga@redhat.com>
11714 Date:   Thu May 9 14:07:48 2019 +0530
11715 
11716     tests/shd: Add test coverage for shd mux
11717     
11718     This patch add more test cases for shd mux test cases
11719     The test case includes
11720     1) Createing multiple volumes to check the attach and detach
11721        of self heal daemon requests.
11722     2) Make sure the healing happens in all sceanarios
11723     3) After a volume detach make sure the threads of the detached
11724        volume is all cleaned.
11725     4) Repeat all the above tests for ec volume
11726     5) Node Reboot case
11727     6) glusterd restart cases
11728     7) Add-brick/remove brick
11729     8) Convert a distributed volume to disperse volume
11730     9) Convert a replicated volume to distributed volume
11731     
11732     Change-Id: I7c317ef9d23a45ffd831157e4890d7c83a8fce7b
11733     fixes: bz#1708929
11734     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
11735 
11736 commit 76673e983b7eb849011b3fb4417cf1d5f7147352
11737 Author: Kotresh HR <khiremat@redhat.com>
11738 Date:   Tue Jun 4 15:10:39 2019 +0530
11739 
11740     tests/geo-rep: Add geo-rep cli testcases
11741     
11742     Change-Id: Icf93b90bcac022a355d4718220698987dbc91ecf
11743     Signed-off-by: Kotresh HR <khiremat@redhat.com>
11744     updates: bz#1693692
11745 
11746 commit 3949d241e8504b1c5226d0e6f6e06a2a12d8357c
11747 Author: N Balachandran <nbalacha@redhat.com>
11748 Date:   Tue Jun 4 14:51:44 2019 +0530
11749 
11750     cluster/dht: Fix directory perms during selfheal
11751     
11752     Fixed a bug in the revalidate code path that wiped out
11753     directory permissions if no mds subvol was found.
11754     
11755     Change-Id: I8b4239ffee7001493c59d4032a2d3062586ea115
11756     fixes: bz#1716830
11757     Signed-off-by: N Balachandran <nbalacha@redhat.com>
11758 
11759 commit e7aeab3063ac5645136303278b477d7de35266c0
11760 Author: Amar Tumballi <amarts@redhat.com>
11761 Date:   Mon May 20 11:11:39 2019 +0530
11762 
11763     across: clang-scan: fix NULL dereferencing warnings
11764     
11765     All these checks are done after analyzing clang-scan report produced
11766     by the CI job @ https://build.gluster.org/job/clang-scan
11767     
11768     updates: bz#1622665
11769     Change-Id: I590305af4ceb779be952974b2a36066ffc4865ca
11770     Signed-off-by: Amar Tumballi <amarts@redhat.com>
11771 
11772 commit abc4d4cfa01d922c46ebb1abfa5d801ba8c6b5c8
11773 Author: Amar Tumballi <amarts@redhat.com>
11774 Date:   Sat Jun 1 22:43:03 2019 +0530
11775 
11776     lcov: run more fops on translators
11777     
11778     Translators covered:
11779     * playground/template
11780     * debug/delay-gen
11781     * debug/error-gen
11782     * features/namespace
11783     * features/quiesce
11784     * meta
11785     
11786     updates: bz#1693692
11787     Change-Id: Ic8fde8efcb309ea492d8e819241f786f7ff467a1
11788     Signed-off-by: Amar Tumballi <amarts@redhat.com>
11789 
11790 commit 400b66d568ad18fefcb59949d1f8368d487b9a80
11791 Author: Krutika Dhananjay <kdhananj@redhat.com>
11792 Date:   Wed May 8 13:00:51 2019 +0530
11793 
11794     features/shard: Fix block-count accounting upon truncate to lower size
11795     
11796     The way delta_blocks is computed in shard is incorrect, when a file
11797     is truncated to a lower size. The accounting only considers change
11798     in size of the last of the truncated shards.
11799     
11800     FIX:
11801     
11802     Get the block-count of each shard just before an unlink at posix in
11803     xdata.  Their summation plus the change in size of last shard
11804     (from an actual truncate) is used to compute delta_blocks which is
11805     used in the xattrop for size update.
11806     
11807     Change-Id: I9128a192e9bf8c3c3a959e96b7400879d03d7c53
11808     fixes: bz#1705884
11809     Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
11810 
11811 commit d8bb432eb776f3a8708ed6cacf1c19fca9524d51
11812 Author: Kotresh HR <khiremat@redhat.com>
11813 Date:   Mon Jun 3 13:25:16 2019 +0530
11814 
11815     tests/geo-rep: Add geo-rep glusterd test cases
11816     
11817     1. Add geo-rep fanout test case
11818     2. Add glusterd geo-rep negative test cases
11819     3. Add glusterd geo-rep config test cases
11820     
11821     Change-Id: I856c087eb3216d8f0ffd1f266deac88e9a4effec
11822     Signed-off-by: Kotresh HR <khiremat@redhat.com>
11823     updates: bz#1693692
11824 
11825 commit 0687befd3332111c7522766aa437ce5166771d0b
11826 Author: Kotresh HR <khiremat@redhat.com>
11827 Date:   Mon Jun 3 14:15:51 2019 +0530
11828 
11829     tests/geo-rep: Remove a rename test case on EC volume
11830     
11831     Rename with existing name testcase is occasionaly
11832     failing on EC volume. Hence commenting the same
11833     until it's analysed
11834     
11835     Change-Id: Icb2ad189b9e4d12101e8f5abcb8a033181360386
11836     Signed-off-by: Kotresh HR <khiremat@redhat.com>
11837     updates: bz#1193929
11838 
11839 commit 823c65940912d7b10bcdd495035dbb23b5c99478
11840 Author: Mohit Agrawal <moagrawal@redhat.com>
11841 Date:   Mon Jun 3 09:49:30 2019 +0530
11842 
11843     glusterd: coverity fix
11844     
11845     1401716: Resource leak
11846     1401714: Dereference before null check
11847     
11848     updates: bz#789278
11849     
11850     Change-Id: I8fb0b143a1d4b37ee6be7d880d9b5b84ba00bf36
11851     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
11852 
11853 commit a36c82375723f51e49c0830d47520392d859c307
11854 Author: hari gowtham <hgowtham@redhat.com>
11855 Date:   Wed May 29 12:26:01 2019 +0530
11856 
11857     glusterd/tier: gluster upgrade broken because of tier
11858     
11859     Problem: While tier code was removed, the is_tier_enabled
11860     related to tier wasn't handled for upgrade.
11861     As this option was missing in the info file, the checksum
11862     mismatch issue happens during upgrade.
11863     This results in the peer rejections happening.
11864     
11865     Fix: use the op_version check and note down the is_tier_enabled
11866     always. This way it will be dummy key, but the future upgrades
11867     will work fine.
11868     
11869     NOTE: Just having the key from 3.10 to 7 will cause issues when
11870     upgraded from 5 to 8 or any such upgrade which skips the version
11871     where we handle it.
11872     
11873     Change-Id: I9951e2b74f16e58e884e746c34dcf53e559c7143
11874     fixes: bz#1714973
11875     Signed-off-by: hari gowtham <hgowtham@redhat.com>
11876 
11877 commit 0f29cabac22c48a0895e2b136f3fd48d9577052b
11878 Author: Amar Tumballi <amarts@redhat.com>
11879 Date:   Thu May 30 15:25:01 2019 +0530
11880 
11881     lcov: improve line coverage
11882     
11883     upcall: remove extra variable assignment and use just one
11884             initialization.
11885     open-behind: reduce the overall number of lines, in functions
11886                  not frequently called
11887     selinux: reduce some lines in init failure cases
11888     
11889     updates: bz#1693692
11890     Change-Id: I7c1de94f2ec76a5bfe1f48a9632879b18e5fbb95
11891     Signed-off-by: Amar Tumballi <amarts@redhat.com>
11892 
11893 commit 899b2170945c6023b0037fed70b19aa3cc680a22
11894 Author: Amar Tumballi <amarts@redhat.com>
11895 Date:   Fri May 17 11:04:45 2019 +0530
11896 
11897     across: coverity fixes
11898     
11899     * locks/posix.c: key was not freed in one of the cases.
11900     * locks/common.c: lock was being free'd out of context.
11901     * nfs/exports: handle case of missing free.
11902     * protocol/client: handle case of entry not freed.
11903     * storage/posix: handle possible case of double free
11904     
11905     CID: 1398628, 1400731, 1400732, 1400756, 1124796, 1325526
11906     
11907     updates: bz#789278
11908     Change-Id: Ieeaca890288bc4686355f6565f853dc8911344e8
11909     Signed-off-by: Amar Tumballi <amarts@redhat.com>
11910     Signed-off-by: Sheetal Pamecha <spamecha@redhat.com>
11911 
11912 commit 950726dfc8e3171bef625b563c0c6dbba1ec2928
11913 Author: Sheetal Pamecha <sheetal.pamecha08@gmail.com>
11914 Date:   Mon Nov 19 22:15:25 2018 +0530
11915 
11916     posix: add storage.reserve-size option
11917     
11918     storage.reserve-size option will take size as input
11919     instead of percentage. If set, priority will be given to
11920     storage.reserve-size over storage.reserve. Default value
11921     of this option is 0.
11922     
11923     fixes: bz#1651445
11924     Change-Id: I7a7342c68e436e8bf65bd39c567512ee04abbcea
11925     Signed-off-by: Sheetal Pamecha <sheetal.pamecha08@gmail.com>
11926 
11927 commit 6fd8281ac9af58609979f660ece58c2ed1100e72
11928 Author: Sanju Rakonde <srakonde@redhat.com>
11929 Date:   Sat Jun 1 18:36:55 2019 +0530
11930 
11931     glusterd: remove trivial conditions
11932     
11933     updates: bz#1193929
11934     
11935     Change-Id: Ieb5e35d454498bc389972f9f15fe46b640f1b97d
11936     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
11937 
11938 commit f8f09178bb890924a8050b466cc2e7a0a30e35a7
11939 Author: Mohit Agrawal <moagrawal@redhat.com>
11940 Date:   Fri May 17 19:26:48 2019 +0530
11941 
11942     glusterd: Optimize code to copy dictionary in handshake code path
11943     
11944     Problem: While high no. of volumes are configured around 2000
11945              glusterd has bottleneck during handshake at the time
11946              of copying dictionary
11947     
11948     Solution: To avoid the bottleneck serialize a dictionary instead
11949               of copying key-value pair one by one
11950     
11951     Change-Id: I9fb332f432e4f915bc3af8dcab38bed26bda2b9a
11952     fixes: bz#1711297
11953     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
11954 
11955 commit fc49927b5fe06380cab309787d2bd62601fa75c4
11956 Author: Kotresh HR <khiremat@redhat.com>
11957 Date:   Fri May 31 16:52:04 2019 +0530
11958 
11959     tests/geo-rep: Add tests to cover glusterd geo-rep
11960     
11961     Change-Id: Ide59a3fde11b23f654b1ec03d72b4ec53b36a03b
11962     Signed-off-by: Kotresh HR <khiremat@redhat.com>
11963     updates: bz#1693692
11964 
11965 commit c848f7163c5ba6b8a081c0f07a4776f7ff8dfca0
11966 Author: Mohammed Rafi KC <rkavunga@redhat.com>
11967 Date:   Tue May 14 23:12:44 2019 +0530
11968 
11969     glusterd/shd: Optimize the glustershd manager to send reconfigure
11970     
11971     Traditionally all svc manager will execute process stop and then
11972     followed by start each time when they called. But that is not
11973     required by shd, because the attach request implemented in the shd
11974     multiplex has the intelligence to check whether a detach is required
11975     prior to attaching the graph. So there is no need to send an explicit
11976     detach request if we are sure that the next call is an attach request
11977     
11978     Change-Id: I9157c8dcaffdac038f73286bcf5646a3f1d3d8ec
11979     fixes: bz#1710054
11980     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
11981 
11982 commit e510f55bed6c26c6b995e7d9f3c35e1e4e482769
11983 Author: Mohammed Rafi KC <rkavunga@redhat.com>
11984 Date:   Sat May 11 22:40:22 2019 +0530
11985 
11986     glusterfsd/cleanup: Protect graph object under a lock
11987     
11988     While processing a cleanup_and_exit function, we are
11989     accessing a graph object. But this has not been protected
11990     under a lock. Because a parallel cleanup of a graph is quite
11991     possible which might lead to an invalid memory access
11992     
11993     Change-Id: Id05ca70d5b57e172b0401d07b6a1f5386c044e79
11994     fixes: bz#1708926
11995     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
11996 
11997 commit f8b0ba06c4cea428f6eb7730f4430b3564c83da0
11998 Author: Shwetha K Acharya <sacharya@redhat.com>
11999 Date:   Fri May 17 12:24:50 2019 +0530
12000 
12001     tests/geo-rep: Add EC volume test case
12002     
12003     Added geo-rep regression tests with EC volume.
12004     
12005     fixes: bz#1650095
12006     Change-Id: Ifb6e68e0a6103a98fced7f84d3088b8edf33d52f
12007     Signed-off-by: Shwetha K Acharya <sacharya@redhat.com>
12008 
12009 commit 4b8244df9d0851ae82cfef5cce359233af7812c8
12010 Author: Amar Tumballi <amarts@redhat.com>
12011 Date:   Fri May 31 08:53:58 2019 +0530
12012 
12013     lcov: more coverage to shard, old-protocol, sdfs
12014     
12015     updates: bz#1693692
12016     Change-Id: If4c30572d4501d169bb4b0871c677d974515867c
12017     Signed-off-by: Amar Tumballi <amarts@redhat.com>
12018 
12019 commit 5866f5e9b09bc10dfb1631b5636c5b7870fc61ff
12020 Author: Mohammed Rafi KC <rkavunga@redhat.com>
12021 Date:   Thu May 30 23:48:05 2019 +0530
12022 
12023     glusterd/svc: Stop stale process using the glusterd_proc_stop
12024     
12025     While restarting a glusterd process, when we have a stale pid
12026     we were doing a simple kill. Instead we can use glusterd_proc_stop
12027     Because it has more logging plus force kill in case if there is
12028     any problem with kill signal handling.
12029     
12030     Change-Id: I4a2dadc210a7a65762dd714e809899510622b7ec
12031     updates: bz#1710054
12032     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
12033 
12034 commit e9ec61c9fc264c813bb9ed06891e2294870a50bc
12035 Author: Mohammed Rafi KC <rkavunga@redhat.com>
12036 Date:   Wed May 22 13:32:23 2019 +0530
12037 
12038     glusterd/svc: glusterd_svcs_stop should call individual wrapper function
12039     
12040     glusterd_svcs_stop should call individual wrapper function to stop a
12041     daemon rather than calling glusterd_svc_stop. For example for shd,
12042     it should call glusterd_shdsvc_stop instead of calling basic API
12043     function to stop. Because the individual functions for each daemon
12044     could be doing some specific operation in their wrapper function.
12045     
12046     Change-Id: Ie6d40590251ad470ef3901d1141ab7b22c3498f5
12047     fixes: bz#1712741
12048     Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com>
12049 
12050 commit 4f1b762fb482f2ebddaacfd31a7d43a967fea9e3
12051 Author: Sanju Rakonde <srakonde@redhat.com>
12052 Date:   Wed May 15 07:35:45 2019 +0530
12053 
12054     glusterd: add an op-version check
12055     
12056     Problem: "gluster v status" is hung in heterogenous cluster
12057     when issued from a non-upgraded node.
12058     
12059     Cause: commit 34e010d64 fixes the txn-opinfo mem leak
12060     in op-sm framework by not setting the txn-opinfo if some
12061     conditions are true. When vol status is issued from a
12062     non-upgraded node, command is hanging in its upgraded peer
12063     as the upgraded node setting the txn-opinfo based on new
12064     conditions where as non-upgraded nodes are following diff
12065     conditions.
12066     
12067     Fix: Add an op-version check, so that all the nodes follow
12068     same set of conditions to set txn-opinfo.
12069     
12070     fixes: bz#1710159
12071     Change-Id: Ie1f353212c5931ddd1b728d2e6949dfe6225c4ab
12072     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
12073 
12074 commit 2855c6247f7c5a2fafe1c403a112c38eca80d020
12075 Author: Pranith Kumar K <pkarampu@redhat.com>
12076 Date:   Mon May 27 19:00:20 2019 +0530
12077 
12078     scripts: Find hung frames given a directory with statedumps
12079     
12080     Given a directory with statedumps captured at different times if
12081     there are any stacks that appear in multiple statedumps, it prints
12082     them.
12083     
12084     Sample output:
12085     glusterdump.25425.dump  repeats=5  stack=0x7f53642cb968  pid=0      unique=0  lk-owner=
12086     glusterdump.25427.dump  repeats=5  stack=0x7f85002cb968  pid=0      unique=0  lk-owner=
12087     glusterdump.25428.dump  repeats=5  stack=0x7f962c2cb968  pid=0      unique=0  lk-owner=
12088     glusterdump.25428.dump  repeats=2  stack=0x7f962c329f18  pid=60830  unique=0  lk-owner=88f50620967f0000
12089     glusterdump.25429.dump  repeats=5  stack=0x7f20782cb968  pid=0      unique=0  lk-owner=
12090     glusterdump.25472.dump  repeats=5  stack=0x7f27ac2cb968  pid=0      unique=0  lk-owner=
12091     glusterdump.25473.dump  repeats=5  stack=0x7f4fbc2cb9d8  pid=0      unique=0  lk-owner=
12092     NOTE: stacks with lk-owner=""/lk-owner=0000000000000000/unique=0 may not be hung frames and need further inspection
12093     
12094     fixes bz#1714415
12095     Change-Id: Ib64a3fca63f49df2fafedcd4baa57e9b25411b08
12096     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
12097 
12098 commit 14d9c6179715205fb6ac816d4621dd087f5afbb2
12099 Author: Pranith Kumar K <pkarampu@redhat.com>
12100 Date:   Mon May 27 11:43:26 2019 +0530
12101 
12102     stack: Make sure to have unique call-stacks in all cases
12103     
12104     At the moment new stack doesn't populate frame->root->unique in all cases. This
12105     makes it difficult to debug hung frames by examining successive state dumps.
12106     Fuse and server xlators populate it whenever they can, but other xlators won't
12107     be able to assign 'unique' when they need to create a new frame/stack because
12108     they don't know what 'unique' fuse/server xlators already used. What we need is
12109     for unique to be correct. If a stack with same unique is present in successive
12110     statedumps, that means the same operation is still in progress. This makes
12111     'finding hung frames' part of debugging hung frames easier.
12112     
12113     fixes bz#1714098
12114     Change-Id: I3e9a8f6b4111e260106c48a2ac3a41ef29361b9e
12115     Signed-off-by: Pranith Kumar K <pkarampu@redhat.com>
12116 
12117 commit d2576452028db3b817fc5b1f4f924a98bc22c7dc
12118 Author: Sanju Rakonde <srakonde@redhat.com>
12119 Date:   Tue May 28 11:52:46 2019 +0530
12120 
12121     glusterd: coverity fix
12122     
12123     1401590: Deadcode
12124     
12125     updates: bz#789278
12126     
12127     Change-Id: I3aa1d3aa9769e6990f74b6a53e288e788173c5e0
12128     Signed-off-by: Sanju Rakonde <srakonde@redhat.com>
12129 
12130 commit 002f4736cd470c20f427d8c25916abb4b95c9240
12131 Author: Amar Tumballi <amarts@redhat.com>
12132 Date:   Wed May 29 17:54:45 2019 +0530
12133 
12134     marker: remove some unused functions
12135     
12136     After basic analysis, found that these methods were not being
12137     used at all.
12138     
12139     updates: bz#1693692
12140     Change-Id: If9cfa1ab189e6e7b56230c4e1d8e11f9694a9a65
12141     Signed-off-by: Amar Tumballi <amarts@redhat.com>
12142 
12143 commit 71caf1423fe5f0eed652210f8158c95fc8a10219
12144 Author: Amar Tumballi <amarts@redhat.com>
12145 Date:   Fri Apr 12 08:16:56 2019 +0530
12146 
12147     tests: add tests for different signal handling
12148     
12149     Also some cleanup:
12150     * old-protocol.t was actually added to make sure we have line-coverage
12151     * first-test.t should have been removed as per the comment. It doesn't do anything.
12152     * add statvfs to rpc-coverage so we can cover statvfs in few xlators.
12153     
12154     updates: bz#1693692
12155     Change-Id: Ie8651ce007de484c4abced16b4de765aa5e517be
12156     Signed-off-by: Amar Tumballi <amarts@redhat.com>
12157 
12158 commit bfe58919c03ea746aa6c7800fe698b15ca8c6ecb
12159 Author: Amgad Saleh <amgad.saleh@nokia.com>
12160 Date:   Thu May 23 20:27:03 2019 -0500
12161 
12162     If bind-address is IPv6 return it successfully
12163     
12164     Change-Id: Ibd37b6ea82b781a1a266b95f7596874134f30079
12165     fixes: bz#1713730
12166     Signed-off-by: Amgad Saleh <amgad.saleh@nokia.com>
12167 
12168 commit 4a5fb52eb1c5387a0fb8bfa1253e5227c7c255e8
12169 Author: Mohit Agrawal <moagrawal@redhat.com>
12170 Date:   Fri May 17 18:10:47 2019 +0530
12171 
12172     glusterd: bulkvoldict thread is not handling all volumes
12173     
12174     Problem: In commit ac70f66c5805e10b3a1072bd467918730c0aeeb4 I
12175              missed one condition to populate volume dictionary in
12176              multiple threads while brick_multiplex is enabled.Due
12177              to that glusterd is not sending volume dictionary for
12178              all volumes to peer.
12179     
12180     Solution: Update the condition in code as well as update test case
12181               also to avoid the issue
12182     
12183     Change-Id: I06522dbdfee4f7e995d9cc7b7098fdf35340dc52
12184     fixes: bz#1711250
12185     Signed-off-by: Mohit Agrawal <moagrawal@redhat.com>
12186 
12187 commit a9d76758454bf7d9a8a3f70fc47b732bb9f97d9e
12188 Author: Kotresh HR <khiremat@redhat.com>
12189 Date:   Wed May 22 12:23:07 2019 +0530
12190 
12191     tests: Add changelog api tests
12192     
12193     updates: bz#1193929
12194     Change-Id: Iee9aab8140882069165621189741f189fb2cc884
12195     Signed-off-by: Kotresh HR <khiremat@redhat.com>
12196 
12197 commit e1cc4275583dfd8ae8d0433587f39854c1851794
12198 Author: Hari Gowtham <hgowtham@redhat.com>
12199 Date:   Thu May 2 18:33:34 2019 +0530
12200 
12201     glusterd/tier: remove tier related code from glusterd
12202     
12203     The handler functions are pointed to dummy functions.
12204     The switch case handling for tier also have been moved to
12205     point default case to avoid issues, if reintroduced.
12206     
12207     The tier changes in DHT still remain as such.
12208     
12209     updates: bz#1693692
12210     
12211     Change-Id: I80d80c9a3eb862b4440a36b31ae82b2e9d92e4dc
12212     Signed-off-by: Hari Gowtham <hgowtham@redhat.com>
12213 
12214 commit cce775726af28c602d99f73a65439a2564790381
12215 Author: Kotresh HR <khiremat@redhat.com>
12216 Date:   Tue May 21 15:28:10 2019 +0530
12217 
12218     tests: Add history api tests
12219     
12220     updates: bz#1193929
12221     Change-Id: Ic26ab5277f720c734f083150c1c541763dfa64aa
12222     Signed-off-by: Kotresh HR <khiremat@redhat.com>
12223 
12224 commit a5a9ea6fa440ebbaeb40cf17d33d39fc3fd6b312
12225 Author: Sheetal Pamecha <spamecha@redhat.com>
12226 Date:   Tue May 7 09:55:08 2019 +0530
12227 
12228     gfapi:add missng api to increase code coverage
12229     
12230     add test for async Read/Write combinations
12231     glfs_read_async/write_async
12232     glfs_pread_async/pwrite_async
12233     glfs_readv_async/writev_async
12234     glfs_preadv_async/pwritev_async
12235     ftruncate/ftruncate_async
12236     fsync/fsync_async
12237     fdatasync/fdatasync_async
12238     
12239     Updates: #655
12240     Change-Id: I12beb97029fd60bce79650a376d8fcd8d383ef16
12241     Signed-off-by: Sheetal Pamecha <spamecha@redhat.com>
12242 
12243 commit 1587a6730ee638d1737b44adbd779d8d2978707d
12244 Author: Sheetal Pamecha <spamecha@redhat.com>
12245 Date:   Mon Apr 15 14:59:37 2019 +0530
12246 
12247     api/glfsxmp.c: minor fixes
12248     
12249     * add more fops: f{get,set,list,remove}xattr(), access(), fstat(), fsetattr(),
12250     getxattr(), lgetxattr(), llistxattr(), lsetxattr(), fgetxattr()
12251     * handle some error cases (like volume not found)
12252     
12253     Updates: #655
12254     Change-Id: I3334bdf3090eafd83a54e1be12036ea01b181089
12255     Signed-off-by: Amar Tumballi <amarts@redhat.com>
12256     Signed-off-by: Sheetal Pamecha <spamecha@redhat.com>
12257 
12258 commit 5d88111a142b3c37e92bdd36699a04fd054d27f4
12259 Author: Xavi Hernandez <xhernandez@redhat.com>
12260 Date:   Wed May 22 17:46:19 2019 +0200
12261 
12262     Fix some "Null pointer dereference" coverity issues
12263     
12264     This patch fixes the following CID's:
12265     
12266       * 1124829
12267       * 1274075
12268       * 1274083
12269       * 1274128
12270       * 1274135
12271       * 1274141
12272       * 1274143
12273       * 1274197
12274       * 1274205
12275       * 1274210
12276       * 1274211
12277       * 1288801
12278       * 1398629
12279     
12280     Change-Id: Ia7c86cfab3245b20777ffa296e1a59748040f558
12281     Updates: bz#789278
12282     Signed-off-by: Xavi Hernandez <xhernandez@redhat.com>
12283 
12284 commit 4c85456e38b9cc2d9698decca15a21e0cb7961a9
12285 Author: Xavi Hernandez <xhernandez@redhat.com>
12286 Date:   Thu May 9 11:07:18 2019 +0200
12287 
12288     cluster/ec: honor contention notifications for partially acquired locks
12289     
12290     EC was ignoring lock contention notifications received while a lock was
12291     being acquired. When a lock is partially acquired (some bricks have
12292     granted the lock but some others not yet) we can receive notifications
12293     from acquired bricks, which should be honored, since we may not receive
12294     more notifications after that.