"Fossies" - the Fresh Open Source Software Archive
Member "udns-0.4/NOTES" (23 May 2011, 11470 Bytes) of package /linux/misc/dns/udns-0.4.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
the uninterpreted source code file.
1 Assorted notes about udns (library).
3 UDP-only mode
6 First of all, since udns is (currently) UDP-only, there are some
9 It assumes that a reply will fit into a UDP buffer. With adoption of EDNS0,
10 and general robustness of IP stacks, in most cases it's not an issue. But
11 in some cases there may be problems:
13 - if an RRset is "very large" so it does not fit even in buffer of size
14 requested by the library (current default is 4096; some servers limits
15 it further), we will not see the reply, or will only see "damaged"
16 reply (depending on the server).
18 - many DNS servers ignores EDNS0 option requests. In this case, no matter
19 which buffer size udns library will request, such servers reply is limited
20 to 512 bytes (standard pre-EDNS0 DNS packet size). (Udns falls back to
21 non-EDNO0 query if EDNS0-enabled one received FORMERR or NOTIMPL error).
23 The problem is that with this, udns currently will not consider replies with
24 TC (truncation) bit set, and will treat such replies the same way as it
25 treats SERVFAIL replies, thus trying next server, or temp-failing the query
26 if no more servers to try. In other words, if the reply is really large, or
27 if the servers you're using don't support EDNS0, your application will be
28 unable to resolve a given name.
30 Yet it's not common situation - in practice, it's very rare.
32 Implementing TCP mode isn't difficult, but it complicates API significantly.
33 Currently udns uses only single UDP socket (or - maybe in the future - two,
34 see below), but in case of TCP, it will need to open and close sockets for
35 TCP connections left and right, and that have to be integrated into an
36 application's event loop in an easy and efficient way. Plus all the
37 timeouts - different for connect(), write, and several stages of read.
39 IPv6 vs IPv4 usage
42 This is only relevant for nameservers reachable over IPv6, NOT for IPv6
43 queries. I.e., if you've IPv6 addresses in 'nameservers' line in your
44 /etc/resolv.conf file. Even more: if you have BOTH IPv6 AND IPv4 addresses
45 there. Or pass them to udns initialization routines.
47 Since udns uses a single UDP socket to communicate with all nameservers,
48 it should support both v4 and v6 communications. Most current platforms
49 supports this mode - using PF_INET6 socket and V4MAPPED addresses, i.e,
50 "tunnelling" IPv4 inside IPv6. But not all systems supports this. And
51 more, it has been said that such mode is deprecated.
53 So, list only IPv4 or only IPv6 addresses, but don't mix them, in your
56 An alternative is to use two sockets instead of 1 - one for IPv6 and one
57 for IPv4. For now I'm not sure if it's worth the complexity - again, of
58 the API, not the library itself (but this will not simplify library either).
60 Single socket for all queries
63 Using single UDP socket for sending queries to all nameservers has obvious
64 advantages. First it's, again, trivial, simple to use API. And simple
65 library too. Also, after sending queries to all nameservers (in case first
66 didn't reply in time), we will be able to receive late reply from first
67 nameserver and accept it.
69 But this mode has disadvantages too. Most important is that it's much easier
70 to send fake reply to us, as the UDP port where we expects the reply to come
71 to is constant during the whole lifetime of an application. More secure
72 implementations uses random port for every single query. While port number
73 (16 bits integer) can not hold much randomness, it's still of some help.
74 Ok, udns is a stub resolver, so it expects sorta friendly environment, but
75 on LAN it's usually much easier to fire an attack, due to the speed of local
76 network, where a bad guy can generate alot of packets in a short time.
78 Spoofing of replies (Kaminsky attack, CVE-2008-1447)
81 While udns uses random numbers for query IDs, it uses single UDP port for
82 all queries (see previous item). And even if it used random UDP port for
83 each query, the attack described in CVE-2008-1447 is still quite trivial.
84 This is not specific to udns library unfortunately - it is inherent property
85 of the protocol. Udns is designed to work in a LAN, it needs full recursive
86 resolver nearby, and modern LAN usually uses high-bandwidth equipment which
87 makes the Kaminsky attack trivial. The problem is that even with qID (16
88 bits) and random UDP port (about 20 bits available to a regular process)
89 combined still can not hold enough randomness, so on a fast network it is
90 still easy to flood the target with fake replies and hit the "right" reply
91 before real reply comes. So random qIDs don't add much protection anyway,
92 even if this feature is implemented in udns, and using all available
93 techniques wont solve it either.
95 See also long comment in udns_resolver.c, udns_newid().
97 Assumptions about RRs returned
100 Currently udns processes records in the reply it received sequentially.
101 This means that order of the records is significant. For example, if
102 we asked for foo.bar A, but the server returned that foo.bar is a CNAME
103 (alias) for bar.baz, and bar.baz, in turn, has address 188.8.131.52, when
104 the CNAME should come first in reply, followed by A. While DNS specs
105 does not say anything about order of records - it's an rrSET - unordered, -
106 I think an implementation which returns the records in "wrong" order is
107 somewhat insane...
109 CNAME recursion
112 Another interesting point is the handling of CNAMEs returned as replies
113 to non-CNAME queries. If we asked for foo.bar A, but it's a CNAME, udns
114 expects BOTH the CNAME itself and the target DN to be present in the reply.
115 In other words, udns DOES NOT RECURSE CNAMES. If we asked for foo.bar A,
116 but only record in reply was that foo.bar is a CNAME for bar.baz, udns will
117 return no records to an application (NXDOMAIN). Strictly speaking, udns
118 should repeat the query asking for bar.baz A, and recurse. But since it's
119 stub resolver, recursive resolver should recurse for us instead.
121 It's not very difficult to implement, however. Probably with some (global?)
122 flag to en/dis-able the feature. Provided there's some demand for it.
124 To clarify: udns handles CNAME recursion in a single reply packet just fine.
126 Note also that standard gethostbyname() routine does not recurse in this
127 situation, too.
129 Error reporting
132 Too many places in the code (various failure paths) sets generic "TEMPFAIL"
133 error condition. For example, if no nameserver replied to our query, an
134 application will get generic TEMPFAIL, instead of something like TIMEDOUT.
135 This probably should be fixed, but most applications don't care about the
136 exact reasons of failure - 4 common cases are already too much:
137 - query returned some valid data
138 - NXDOMAIN
139 - valid domain but no data of requested type - =NXDOMAIN in most cases
140 - temporary error - this one sometimes (incorrectly!) treated as NXDOMAIN
141 by (naive) applications.
142 DNS isn't yes/no, it's at least 3 variants, temp err being the 3rd important
143 case! And adding more variations for the temp error case is complicating things
144 even more - again, from an application writer standpoint. For diagnostics,
145 such more specific error cases are of good help.
147 Planned API changes
150 At least one thing I want to change for some future version is a way how
151 queries are submitted and how replies are handled.
153 I want to made dns_query object to be owned by an application. So that instead
154 of udns library allocating it for the lifetime of query, it will be pre-
155 allocated by an application. This simplifies and enhances query submitting
156 interface, and complicates it a bit too, in simplest cases.
158 Currently, we have:
160 dns_submit_dn(dn, cls, typ, flags, parse, cbck, data)
161 dns_submit_p(name, cls, typ, flags, parse, cbck, data)
162 dns_submit_a4(ctx, name, flags, cbck, data)
164 and so on -- with many parameters missed for type-specific cases, but generic
165 cases being too complex for most common usage.
167 Instead, with dns_query being owned by an app, we will be able to separately
168 set up various parts of the query - domain name (various forms), type&class,
169 parser, flags, callback... and even change them at runtime. And we will also
170 be able to reuse query structures, instead of allocating/freeing them every
171 time. So the whole thing will look something like:
173 q = dns_alloc_query();
174 dns_submit(dns_q_flags(dns_q_a4(q, name, cbck), DNS_F_NOSRCH), data);
176 The idea is to have a set of functions accepting struct dns_query* and
177 returning it (so the calls can be "nested" like the above), to set up
178 relevant parts of the query - specific type of callback, conversion from
179 (type-specific) query parameters into a domain name (this is for type-
180 specific query initializers), and setting various flags and options and
181 type&class things.
183 One example where this is almost essential - if we want to support
184 per-query set of nameservers (which isn't at all useless: imagine a
185 high-volume mail server, were we want to direct DNSBL queries to a separate
186 set of nameservers, and rDNS queries to their own set and so on). Adding
187 another argument (set of nameservers to use) to EVERY query submitting
188 routine is.. insane. Especially since in 99% cases it will be set to
189 default NULL. But with such "nesting" of query initializers, it becomes
192 This change (the way how queries gets submitted) will NOT break API/ABI
193 compatibility with old versions, since the new submitting API works in
194 parallel with current (and current will use the new one as building
195 blocks, instead of doing all work at once).
197 Another way to do the same is to manipulate query object right after a
198 query has been submitted, but before any events processing (during this
199 time, query object is allocated and initialized, but no actual network
200 packets were sent - it will happen on the next event processing). But
201 this way it become impossible to perform syncronous resolver calls, since
202 those calls hide query objects they use internally.
204 Speaking of replies handling - the planned change is to stop using dynamic
205 memory (malloc) inside the library. That is, instead of allocating a buffer
206 for a reply dynamically in a parsing routine (or memdup'ing the raw reply
207 packet if no parsing routine is specified), I want udns to return the packet
208 buffer it uses internally, and change parsing routines to expect a buffer
209 for result. When parsing, a routine will return true amount of memory it
210 will need to place the result, regardless of whenever it has enough room
211 or not, so that an application can (re)allocate properly sized buffer and
212 call a parsing routine again.
214 This, in theory, also can be done without breaking current API/ABI, but in
215 that case we'll again need a parallel set of routines (parsing included),
216 which makes the library more complicated with too many ways of doing the
217 same thing. Still, code reuse is at good level.
219 Another modification I plan to include is to have an ability to work in
220 terms of domain names (DNs) as used with on-wire DNS packets, not only
221 with asciiz representations of them. For this to work, the above two
222 changes (query submission and result passing) have to be completed first
223 (esp. the query submission part), so that it will be possible to specify
224 some additional query flags (for example) to request domain names instead
225 of the text strings, and to allow easy query submissions with either DNs
226 or text strings.