7Network Working Group R. Austein
8Request for Comments: 5001 ISC
9Category: Standards Track August 2007
12 DNS Name Server Identifier (NSID) Option
16 This document specifies an Internet standards track protocol for the
17 Internet community, and requests discussion and suggestions for
18 improvements. Please refer to the current edition of the "Internet
19 Official Protocol Standards" (STD 1) for the standardization state
20 and status of this protocol. Distribution of this memo is unlimited.
24 Copyright (C) The IETF Trust (2007).
28 With the increased use of DNS anycast, load balancing, and other
29 mechanisms allowing more than one DNS name server to share a single
30 IP address, it is sometimes difficult to tell which of a pool of name
31 servers has answered a particular query. While existing ad-hoc
32 mechanisms allow an operator to send follow-up queries when it is
33 necessary to debug such a configuration, the only completely reliable
34 way to obtain the identity of the name server that responded is to
35 have the name server include this information in the response itself.
36 This note defines a protocol extension to support this functionality.
58Austein Standards Track [Page 1]
60RFC 5001 DNS NSID August 2007
65 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 2
66 1.1. Reserved Words . . . . . . . . . . . . . . . . . . . . . . 3
67 2. Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
68 2.1. Resolver Behavior . . . . . . . . . . . . . . . . . . . . 3
69 2.2. Name Server Behavior . . . . . . . . . . . . . . . . . . . 3
70 2.3. The NSID Option . . . . . . . . . . . . . . . . . . . . . 4
71 2.4. Presentation Format . . . . . . . . . . . . . . . . . . . 4
72 3. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . 4
73 3.1. The NSID Payload . . . . . . . . . . . . . . . . . . . . . 4
74 3.2. NSID Is Not Transitive . . . . . . . . . . . . . . . . . . 7
75 3.3. User Interface Issues . . . . . . . . . . . . . . . . . . 7
76 3.4. Truncation . . . . . . . . . . . . . . . . . . . . . . . . 8
77 4. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 8
78 5. Security Considerations . . . . . . . . . . . . . . . . . . . 9
79 6. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 9
80 7. References . . . . . . . . . . . . . . . . . . . . . . . . . . 9
81 7.1. Normative References . . . . . . . . . . . . . . . . . . . 9
82 7.2. Informative References . . . . . . . . . . . . . . . . . . 10
86 With the increased use of DNS anycast, load balancing, and other
87 mechanisms allowing more than one DNS name server to share a single
88 IP address, it is sometimes difficult to tell which of a pool of name
89 servers has answered a particular query.
91 Existing ad-hoc mechanisms allow an operator to send follow-up
92 queries when it is necessary to debug such a configuration, but there
93 are situations in which this is not a totally satisfactory solution,
94 since anycast routing may have changed, or the server pool in
95 question may be behind some kind of extremely dynamic load balancing
96 hardware. Thus, while these ad-hoc mechanisms are certainly better
97 than nothing (and have the advantage of already being deployed), a
98 better solution seems desirable.
100 Given that a DNS query is an idempotent operation with no retained
101 state, it would appear that the only completely reliable way to
102 obtain the identity of the name server that responded to a particular
103 query is to have that name server include identifying information in
104 the response itself. This note defines a protocol enhancement to
114Austein Standards Track [Page 2]
116RFC 5001 DNS NSID August 2007
121 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
122 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
123 document are to be interpreted as described in [RFC2119].
127 This note uses an EDNS [RFC2671] option to signal the resolver's
128 desire for information identifying the name server and to hold the
129 name server's response, if any.
1312.1. Resolver Behavior
133 A resolver signals its desire for information identifying a name
134 server by sending an empty NSID option (Section 2.3) in an EDNS OPT
135 pseudo-RR in the query message.
137 The resolver MUST NOT include any NSID payload data in the query
140 The semantics of an NSID request are not transitive. That is: the
141 presence of an NSID option in a query is a request that the name
142 server which receives the query identify itself. If the name server
143 side of a recursive name server receives an NSID request, the client
144 is asking the recursive name server to identify itself; if the
145 resolver side of the recursive name server wishes to receive
146 identifying information, it is free to add NSID requests in its own
147 queries, but that is a separate matter.
1492.2. Name Server Behavior
151 A name server that understands the NSID option and chooses to honor a
152 particular NSID request responds by including identifying information
153 in a NSID option (Section 2.3) in an EDNS OPT pseudo-RR in the
156 The name server MUST ignore any NSID payload data that might be
157 present in the query message.
159 The NSID option is not transitive. A name server MUST NOT send an
160 NSID option back to a resolver which did not request it. In
161 particular, while a recursive name server may choose to add an NSID
162 option when sending a query, this has no effect on the presence or
163 absence of the NSID option in the recursive name server's response to
170Austein Standards Track [Page 3]
172RFC 5001 DNS NSID August 2007
175 As stated in Section 2.1, this mechanism is not restricted to
176 authoritative name servers; the semantics are intended to be equally
177 applicable to recursive name servers.
181 The OPTION-CODE for the NSID option is 3.
183 The OPTION-DATA for the NSID option is an opaque byte string, the
184 semantics of which are deliberately left outside the protocol. See
185 Section 3.1 for discussion.
1872.4. Presentation Format
189 User interfaces MUST read and write the contents of the NSID option
190 as a sequence of hexadecimal digits, two digits per payload octet.
192 The NSID payload is binary data. Any comparison between NSID
193 payloads MUST be a comparison of the raw binary data. Copy
194 operations MUST NOT assume that the raw NSID payload is null-
195 terminated. Any resemblance between raw NSID payload data and any
196 form of text is purely a convenience, and does not change the
197 underlying nature of the payload data.
199 See Section 3.3 for discussion.
203 This section discusses certain aspects of the protocol and explains
204 considerations that led to the chosen design.
208 The syntax and semantics of the content of the NSID option are
209 deliberately left outside the scope of this specification.
211 Choosing the NSID content is a prerogative of the server
212 administrator. The server administrator might choose to encode the
213 NSID content in such a way that the server operator (or clients
214 authorized by the server operator) can decode the NSID content to
215 obtain more information than other clients can. Alternatively, the
216 server operator might choose unencoded NSID content that is equally
217 meaningful to any client.
219 This section describes some of the kinds of data that server
220 administrators might choose to provide as the content of the NSID
221 option, and explains the reasoning behind specifying a simple opaque
222 byte string in Section 2.3.
226Austein Standards Track [Page 4]
228RFC 5001 DNS NSID August 2007
231 There are several possibilities for the payload of the NSID option:
233 o It could be the "real" name of the specific name server within the
236 o It could be the "real" IP address (IPv4 or IPv6) of the name
237 server within the name server pool.
239 o It could be some sort of pseudo-random number generated in a
240 predictable fashion somehow using the server's IP address or name
243 o It could be some sort of probabilistically unique identifier
244 initially derived from some sort of random number generator then
245 preserved across reboots of the name server.
247 o It could be some sort of dynamically generated identifier so that
248 only the name server operator could tell whether or not any two
249 queries had been answered by the same server.
251 o It could be a blob of signed data, with a corresponding key which
252 might (or might not) be available via DNS lookups.
254 o It could be a blob of encrypted data, the key for which could be
255 restricted to parties with a need to know (in the opinion of the
258 o It could be an arbitrary string of octets chosen at the discretion
259 of the name server operator.
261 Each of these options has advantages and disadvantages:
263 o Using the "real" name is simple, but the name server may not have
266 o Using the "real" address is also simple, and the name server
267 almost certainly does have at least one non-anycast IP address for
268 maintenance operations, but the operator of the name server may
269 not be willing to divulge its non-anycast address.
271 o Given that one common reason for using anycast DNS techniques is
272 an attempt to harden a critical name server against denial of
273 service attacks, some name server operators are likely to want an
274 identifier other than the "real" name or "real" address of the
275 name server instance.
277 o Using a hash or pseudo-random number can provide a fixed length
278 value that the resolver can use to tell two name servers apart
282Austein Standards Track [Page 5]
284RFC 5001 DNS NSID August 2007
287 without necessarily being able to tell where either one of them
288 "really" is, but makes debugging more difficult if one happens to
289 be in a friendly open environment. Furthermore, hashing might not
290 add much value, since a hash based on an IPv4 address still only
291 involves a 32-bit search space, and DNS names used for servers
292 that operators might have to debug at 4am tend not to be very
295 o Probabilistically unique identifiers have properties similar to
296 hashed identifiers, but (given a sufficiently good random number
297 generator) are immune to the search space issues. However, the
298 strength of this approach is also its weakness: there is no
299 algorithmic transformation by which even the server operator can
300 associate name server instances with identifiers while debugging,
301 which might be annoying. This approach also requires the name
302 server instance to preserve the probabilistically unique
303 identifier across reboots, but this does not appear to be a
304 serious restriction, since authoritative nameservers almost always
305 have some form of non-volatile storage. In the rare case of a
306 name server that does not have any way to store such an
307 identifier, nothing terrible will happen if the name server
308 generates a new identifier every time it reboots.
310 o Using an arbitrary octet string gives name server operators yet
311 another setting to configure, or mis-configure, or forget to
312 configure. Having all the nodes in an anycast name server
313 constellation identify themselves as "My Name Server" would not be
316 o A signed blob is not particularly useful as an NSID payload unless
317 the signed data is dynamic and includes some kind of replay
318 protection, such as a timestamp or some kind of data identifying
319 the requestor. Signed blobs that meet these criteria could
320 conceivably be useful in some situations but would require
321 detailed security analysis beyond the scope of this document.
323 o A static encrypted blob would not be particularly useful, as it
324 would be subject to replay attacks and would, in effect, just be a
325 random number to any party that does not possess the decryption
326 key. Dynamic encrypted blobs could conceivably be useful in some
327 situations but, as with signed blobs, dynamic encrypted blobs
328 would require detailed security analysis beyond the scope of this
331 Given all of the issues listed above, there does not appear to be a
332 single solution that will meet all needs. Section 2.3 therefore
333 defines the NSID payload to be an opaque byte string and leaves the
334 choice of payload up to the implementor and name server operator.
338Austein Standards Track [Page 6]
340RFC 5001 DNS NSID August 2007
343 The following guidelines may be useful to implementors and server
346 o Operators for whom divulging the unicast address is an issue could
347 use the raw binary representation of a probabilistically unique
348 random number. This should probably be the default implementation
351 o Operators for whom divulging the unicast address is not an issue
352 could just use the raw binary representation of a unicast address
353 for simplicity. This should only be done via an explicit
354 configuration choice by the operator.
356 o Operators who really need or want the ability to set the NSID
357 payload to an arbitrary value could do so, but this should only be
358 done via an explicit configuration choice by the operator.
360 This approach appears to provide enough information for useful
361 debugging without unintentionally leaking the maintenance addresses
362 of anycast name servers to nogoodniks, while also allowing name
363 server operators who do not find such leakage threatening to provide
364 more information at their own discretion.
3663.2. NSID Is Not Transitive
368 As specified in Section 2.1 and Section 2.2, the NSID option is not
369 transitive. This is strictly a hop-by-hop mechanism.
371 Most of the discussion of name server identification to date has
372 focused on identifying authoritative name servers, since the best
373 known cases of anycast name servers are a subset of the name servers
374 for the root zone. However, given that anycast DNS techniques are
375 also applicable to recursive name servers, the mechanism may also be
376 useful with recursive name servers. The hop-by-hop semantics support
379 While there might be some utility in having a transitive variant of
380 this mechanism (so that, for example, a stub resolver could ask a
381 recursive server to tell it which authoritative name server provided
382 a particular answer to the recursive name server), the semantics of
383 such a variant would be more complicated, and are left for future
3863.3. User Interface Issues
388 Given the range of possible payload contents described in
389 Section 3.1, it is not possible to define a single presentation
390 format for the NSID payload that is efficient, convenient,
394Austein Standards Track [Page 7]
396RFC 5001 DNS NSID August 2007
399 unambiguous, and aesthetically pleasing. In particular, while it is
400 tempting to use a presentation format that uses some form of textual
401 strings, attempting to support this would significantly complicate
402 what's intended to be a very simple debugging mechanism.
404 In some cases the content of the NSID payload may be binary data
405 meaningful only to the name server operator, and may not be
406 meaningful to the user or application, but the user or application
407 must be able to capture the entire content anyway in order for it to
408 be useful. Thus, the presentation format must support arbitrary
411 In cases where the name server operator derives the NSID payload from
412 textual data, a textual form such as US-ASCII or UTF-8 strings might
413 at first glance seem easier for a user to deal with. There are,
414 however, a number of complex issues involving internationalized text
415 which, if fully addressed here, would require a set of rules
416 significantly longer than the rest of this specification. See
417 [RFC2277] for an overview of some of these issues.
419 It is much more important for the NSID payload data to be passed
420 unambiguously from server administrator to user and back again than
421 it is for the payload data to be pretty while in transit. In
422 particular, it's critical that it be straightforward for a user to
423 cut and paste an exact copy of the NSID payload output by a debugging
424 tool into other formats such as email messages or web forms without
425 distortion. Hexadecimal strings, while ugly, are also robust.
429 In some cases, adding the NSID option to a response message may
430 trigger message truncation. This specification does not change the
431 rules for DNS message truncation in any way, but implementors will
432 need to pay attention to this issue.
434 Including the NSID option in a response is always optional, so this
435 specification never requires name servers to truncate response
438 By definition, a resolver that requests NSID responses also supports
439 EDNS, so a resolver that requests NSID responses can also use the
440 "sender's UDP payload size" field of the OPT pseudo-RR to signal a
441 receive buffer size large enough to make truncation unlikely.
4434. IANA Considerations
445 IANA has allocated EDNS option code 3 for the NSID option
450Austein Standards Track [Page 8]
452RFC 5001 DNS NSID August 2007
4555. Security Considerations
457 This document describes a channel signaling mechanism intended
458 primarily for debugging. Channel signaling mechanisms are outside
459 the scope of DNSSEC, per se. Applications that require integrity
460 protection for the data being signaled will need to use a channel
461 security mechanism such as TSIG [RFC2845].
463 Section 3.1 discusses a number of different kinds of information that
464 a name server operator might choose to provide as the value of the
465 NSID option. Some of these kinds of information are security
466 sensitive in some environments. This specification deliberately
467 leaves the syntax and semantics of the NSID option content up to the
468 implementation and the name server operator.
470 Two of the possible kinds of payload data discussed in Section 3.1
471 involve a digital signature and encryption, respectively. While this
472 specification discusses some of the pitfalls that might lurk for
473 careless users of these kinds of payload data, full analysis of the
474 issues that would be involved in these kinds of payload data would
475 require knowledge of the content to be signed or encrypted,
476 algorithms to be used, and so forth, which is beyond the scope of
477 this document. Implementors should seek competent advice before
478 attempting to use these kinds of NSID payloads.
482 Thanks to: Joe Abley, Harald Alvestrand, Dean Anderson, Mark Andrews,
483 Roy Arends, Steve Bellovin, Alex Bligh, Randy Bush, David Conrad,
484 John Dickinson, Alfred Hoenes, Johan Ihren, Daniel Karrenberg, Peter
485 Koch, William Leibzon, Ed Lewis, Thomas Narten, Mike Patton, Geoffrey
486 Sisson, Andrew Sullivan, Mike StJohns, Tom Taylor, Paul Vixie, Sam
487 Weiler, and Suzanne Woolf, none of whom are responsible for what the
488 author did with their comments and suggestions. Apologies to anyone
489 inadvertently omitted from the above list.
4937.1. Normative References
495 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
496 Requirement Levels", RFC 2119, BCP 14, March 1997.
498 [RFC2671] Vixie, P., "Extension Mechanisms for DNS (EDNS0)",
499 RFC 2671, August 1999.
506Austein Standards Track [Page 9]
508RFC 5001 DNS NSID August 2007
511 [RFC2845] Vixie, P., Gudmundsson, O., Eastlake 3rd, D., and B.
512 Wellington, "Secret Key Transaction Authentication for DNS
513 (TSIG)", RFC 2845, May 2000.
5157.2. Informative References
517 [RFC2277] Alvestrand, H., "IETF Policy on Character Sets and
518 Languages", RFC 2277, BCP 18, January 1998.
525 Redwood City, CA 94063
562Austein Standards Track [Page 10]
564RFC 5001 DNS NSID August 2007
567Full Copyright Statement
569 Copyright (C) The IETF Trust (2007).
571 This document is subject to the rights, licenses and restrictions
572 contained in BCP 78, and except as set forth therein, the authors
573 retain all their rights.
575 This document and the information contained herein are provided on an
576 "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS
577 OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND
578 THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS
579 OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF
580 THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
581 WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
585 The IETF takes no position regarding the validity or scope of any
586 Intellectual Property Rights or other rights that might be claimed to
587 pertain to the implementation or use of the technology described in
588 this document or the extent to which any license under such rights
589 might or might not be available; nor does it represent that it has
590 made any independent effort to identify any such rights. Information
591 on the procedures with respect to rights in RFC documents can be
592 found in BCP 78 and BCP 79.
594 Copies of IPR disclosures made to the IETF Secretariat and any
595 assurances of licenses to be made available, or the result of an
596 attempt made to obtain a general license or permission for the use of
597 such proprietary rights by implementers or users of this
598 specification can be obtained from the IETF on-line IPR repository at
599 http://www.ietf.org/ipr.
601 The IETF invites any interested party to bring to its attention any
602 copyrights, patents or patent applications, or other proprietary
603 rights that may cover technology that may be required to implement
604 this standard. Please address the information to the IETF at
609 Funding for the RFC Editor function is currently provided by the
618Austein Standards Track [Page 11]