4 Network Working Group R. Austein
6 Expires: July 15, 2006 January 11, 2006
9 DNS Name Server Identifier Option (NSID)
10 draft-ietf-dnsext-nsid-01
14 By submitting this Internet-Draft, each author represents that any
15 applicable patent or other IPR claims of which he or she is aware
16 have been or will be disclosed, and any of which he or she becomes
17 aware will be disclosed, in accordance with Section 6 of BCP 79.
19 Internet-Drafts are working documents of the Internet Engineering
20 Task Force (IETF), its areas, and its working groups. Note that
21 other groups may also distribute working documents as Internet-
24 Internet-Drafts are draft documents valid for a maximum of six months
25 and may be updated, replaced, or obsoleted by other documents at any
26 time. It is inappropriate to use Internet-Drafts as reference
27 material or to cite them other than as "work in progress."
29 The list of current Internet-Drafts can be accessed at
30 http://www.ietf.org/ietf/1id-abstracts.txt.
32 The list of Internet-Draft Shadow Directories can be accessed at
33 http://www.ietf.org/shadow.html.
35 This Internet-Draft will expire on July 15, 2006.
39 Copyright (C) The Internet Society (2006).
43 With the increased use of DNS anycast, load balancing, and other
44 mechanisms allowing more than one DNS name server to share a single
45 IP address, it is sometimes difficult to tell which of a pool of name
46 servers has answered a particular query. While existing ad-hoc
47 mechanism allow an operator to send follow-up queries when it is
48 necessary to debug such a configuration, the only completely reliable
49 way to obtain the identity of the name server which responded is to
50 have the name server include this information in the response itself.
51 This note defines a protocol extension to support this functionality.
55 Austein Expires July 15, 2006 [Page 1]
57 Internet-Draft DNS NSID January 2006
62 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3
63 1.1. Reserved Words . . . . . . . . . . . . . . . . . . . . . . 3
64 2. Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
65 2.1. Resolver Behavior . . . . . . . . . . . . . . . . . . . . 4
66 2.2. Name Server Behavior . . . . . . . . . . . . . . . . . . . 4
67 2.3. The NSID Option . . . . . . . . . . . . . . . . . . . . . 4
68 2.4. Presentation Format . . . . . . . . . . . . . . . . . . . 5
69 3. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . 6
70 3.1. The NSID Payload . . . . . . . . . . . . . . . . . . . . . 6
71 3.2. NSID Is Not Transitive . . . . . . . . . . . . . . . . . . 8
72 3.3. User Interface Issues . . . . . . . . . . . . . . . . . . 8
73 3.4. Truncation . . . . . . . . . . . . . . . . . . . . . . . . 9
74 4. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 10
75 5. Security Considerations . . . . . . . . . . . . . . . . . . . 11
76 6. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 12
77 7. References . . . . . . . . . . . . . . . . . . . . . . . . . . 13
78 7.1. Normative References . . . . . . . . . . . . . . . . . . . 13
79 7.2. Informative References . . . . . . . . . . . . . . . . . . 13
80 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . . 14
81 Intellectual Property and Copyright Statements . . . . . . . . . . 15
111 Austein Expires July 15, 2006 [Page 2]
113 Internet-Draft DNS NSID January 2006
118 With the increased use of DNS anycast, load balancing, and other
119 mechanisms allowing more than one DNS name server to share a single
120 IP address, it is sometimes difficult to tell which of a pool of name
121 servers has answered a particular query.
123 Existing ad-hoc mechanisms allow an operator to send follow-up
124 queries when it is necessary to debug such a configuration, but there
125 are situations in which this is not a totally satisfactory solution,
126 since anycast routing may have changed, or the server pool in
127 question may be behind some kind of extremely dynamic load balancing
128 hardware. Thus, while these ad-hoc mechanisms are certainly better
129 than nothing (and have the advantage of already being deployed), a
130 better solution seems desirable.
132 Given that a DNS query is an idempotent operation with no retained
133 state, it would appear that the only completely reliable way to
134 obtain the identity of the name server which responded to a
135 particular query is to have that name server include identifying
136 information in the response itself. This note defines a protocol
137 enhancement to achieve this.
141 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
142 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
143 document are to be interpreted as described in [RFC2119].
167 Austein Expires July 15, 2006 [Page 3]
169 Internet-Draft DNS NSID January 2006
174 This note uses an EDNS [RFC2671] option to signal the resolver's
175 desire for information identifying the name server and to hold the
176 name server's response, if any.
178 2.1. Resolver Behavior
180 A resolver signals its desire for information identifying a name
181 server by sending an empty NSID option (Section 2.3) in an EDNS OPT
182 pseudo-RR in the query message.
184 The resolver MUST NOT include any NSID payload data in the query
187 The semantics of an NSID request are not transitive. That is: the
188 presence of an NSID option in a query is a request that the name
189 server which receives the query identify itself. If the name server
190 side of a recursive name server receives an NSID request, the client
191 is asking the recursive name server to identify itself; if the
192 resolver side of the recursive name server wishes to receive
193 identifying information, it is free to add NSID requests in its own
194 queries, but that is a separate matter.
196 2.2. Name Server Behavior
198 A name server which understands the NSID option and chooses to honor
199 a particular NSID request responds by including identifying
200 information in a NSID option (Section 2.3) in an EDNS OPT pseudo-RR
201 in the response message.
203 The name server MUST ignore any NSID payload data that might be
204 present in the query message.
206 The NSID option is not transitive. A name server MUST NOT send an
207 NSID option back to a resolver which did not request it. In
208 particular, while a recursive name server may choose to add an NSID
209 option when sending a query, this has no effect on the presence or
210 absence of the NSID option in the recursive name server's response to
213 As stated in Section 2.1, this mechanism is not restricted to
214 authoritative name servers; the semantics are intended to be equally
215 applicable to recursive name servers.
219 The OPTION-CODE for the NSID option is [TBD].
223 Austein Expires July 15, 2006 [Page 4]
225 Internet-Draft DNS NSID January 2006
228 The OPTION-DATA for the NSID option is an opaque byte string the
229 semantics of which are deliberately left outside the protocol. See
230 Section 3.1 for discussion.
232 2.4. Presentation Format
234 User interfaces MUST read and write the content of the NSID option as
235 a sequence of hexadecimal digits, two digits per payload octet.
237 The NSID payload is binary data. Any comparison between NSID
238 payloads MUST be a comparison of the raw binary data. Copy
239 operations MUST NOT assume that the raw NSID payload is null-
240 terminated. Any resemblance between raw NSID payload data and any
241 form of text is purely a convenience, and does not change the
242 underlying nature of the payload data.
244 See Section 3.3 for discussion.
279 Austein Expires July 15, 2006 [Page 5]
281 Internet-Draft DNS NSID January 2006
286 This section discusses certain aspects of the protocol and explains
287 considerations that led to the chosen design.
289 3.1. The NSID Payload
291 The syntax and semantics of the content of the NSID option is
292 deliberately left outside the scope of this specification. This
293 section describe some of the kinds of data that server administrators
294 might choose to provide as the content of the NSID option, and
295 explains the reasoning behind choosing a simple opaque byte string.
297 There are several possibilities for the payload of the NSID option:
299 o It could be the "real" name of the specific name server within the
302 o It could be the "real" IP address (IPv4 or IPv6) of the name
303 server within the name server pool.
305 o It could be some sort of pseudo-random number generated in a
306 predictable fashion somehow using the server's IP address or name
309 o It could be some sort of probabilisticly unique identifier
310 initially derived from some sort of random number generator then
311 preserved across reboots of the name server.
313 o It could be some sort of dynamicly generated identifier so that
314 only the name server operator could tell whether or not any two
315 queries had been answered by the same server.
317 o It could be a blob of signed data, with a corresponding key which
318 might (or might not) be available via DNS lookups.
320 o It could be a blob of encrypted data, the key for which could be
321 restricted to parties with a need to know (in the opinion of the
324 o It could be an arbitrary string of octets chosen at the discretion
325 of the name server operator.
327 Each of these options has advantages and disadvantages:
329 o Using the "real" name is simple, but the name server may not have
335 Austein Expires July 15, 2006 [Page 6]
337 Internet-Draft DNS NSID January 2006
340 o Using the "real" address is also simple, and the name server
341 almost certainly does have at least one non-anycast IP address for
342 maintenance operations, but the operator of the name server may
343 not be willing to divulge its non-anycast address.
345 o Given that one common reason for using anycast DNS techniques is
346 an attempt to harden a critical name server against denial of
347 service attacks, some name server operators are likely to want an
348 identifier other than the "real" name or "real" address of the
349 name server instance.
351 o Using a hash or pseudo-random number can provide a fixed length
352 value that the resolver can use to tell two name servers apart
353 without necessarily being able to tell where either one of them
354 "really" is, but makes debugging more difficult if one happens to
355 be in a friendly open environment. Furthermore, hashing might not
356 add much value, since a hash based on an IPv4 address still only
357 involves a 32-bit search space, and DNS names used for servers
358 that operators might have to debug at 4am tend not to be very
361 o Probabilisticly unique identifiers have similar properties to
362 hashed identifiers, but (given a sufficiently good random number
363 generator) are immune to the search space issues. However, the
364 strength of this approach is also its weakness: there is no
365 algorithmic transformation by which even the server operator can
366 associate name server instances with identifiers while debugging,
367 which might be annoying. This approach also requires the name
368 server instance to preserve the probabilisticly unique identifier
369 across reboots, but this does not appear to be a serious
370 restriction, since authoritative nameservers almost always have
371 some form of nonvolatile storage in any case, and in the rare case
372 of a name server that does not have any way to store such an
373 identifier, nothing terrible will happen if the name server just
374 generates a new identifier every time it reboots.
376 o Using an arbitrary octet string gives name server operators yet
377 another thing to configure, or mis-configure, or forget to
378 configure. Having all the nodes in an anycast name server
379 constellation identify themselves as "My Name Server" would not be
382 Given all of the issues listed above, there does not appear to be a
383 single solution that will meet all needs. Section 2.3 therefore
384 defines the NSID payload to be an opaque byte string and leaves the
385 choice up to the implementor and name server operator. The following
386 guidelines may be useful to implementors and server operators:
391 Austein Expires July 15, 2006 [Page 7]
393 Internet-Draft DNS NSID January 2006
396 o Operators for whom divulging the unicast address is an issue could
397 use the raw binary representation of a probabilisticly unique
398 random number. This should probably be the default implementation
401 o Operators for whom divulging the unicast address is not an issue
402 could just use the raw binary representation of a unicast address
403 for simplicity. This should only be done via an explicit
404 configuration choice by the operator.
406 o Operators who really need or want the ability to set the NSID
407 payload to an arbitrary value could do so, but this should only be
408 done via an explicit configuration choice by the operator.
410 This approach appears to provide enough information for useful
411 debugging without unintentionally leaking the maintenance addresses
412 of anycast name servers to nogoodniks, while also allowing name
413 server operators who do not find such leakage threatening to provide
414 more information at their own discretion.
416 3.2. NSID Is Not Transitive
418 As specified in Section 2.1 and Section 2.2, the NSID option is not
419 transitive. This is strictly a hop-by-hop mechanism.
421 Most of the discussion of name server identification to date has
422 focused on identifying authoritative name servers, since the best
423 known cases of anycast name servers are a subset of the name servers
424 for the root zone. However, given that anycast DNS techniques are
425 also applicable to recursive name servers, the mechanism may also be
426 useful with recursive name servers. The hop-by-hop semantics support
429 While there might be some utility in having a transitive variant of
430 this mechanism (so that, for example, a stub resolver could ask a
431 recursive server to tell it which authoritative name server provided
432 a particular answer to the recursive name server), the semantics of
433 such a variant would be more complicated, and are left for future
436 3.3. User Interface Issues
438 Given the range of possible payload contents described in
439 Section 3.1, it is not possible to define a single presentation
440 format for the NSID payload that is efficient, convenient,
441 unambiguous, and aesthetically pleasing. In particular, while it is
442 tempting to use a presentation format that uses some form of textual
443 strings, attempting to support this would significantly complicate
447 Austein Expires July 15, 2006 [Page 8]
449 Internet-Draft DNS NSID January 2006
452 what's intended to be a very simple debugging mechanism.
454 In some cases the content of the NSID payload may be binary data
455 meaningful only to the name server operator, and may not be
456 meaningful to the user or application, but the user or application
457 must be able to capture the entire content anyway in order for it to
458 be useful. Thus, the presentation format must support arbitrary
461 In cases where the name server operator derives the NSID payload from
462 textual data, a textual form such as US-ASCII or UTF-8 strings might
463 at first glance seem easier for a user to deal with. There are,
464 however, a number of complex issues involving internationalized text
465 which, if fully addressed here, would require a set of rules
466 significantly longer than the rest of this specification. See
467 [RFC2277] for an overview of some of these issues.
469 It is much more important for the NSID payload data to be passed
470 unambiguously from server administrator to user and back again than
471 it is for the payload data data to be pretty while in transit. In
472 particular, it's critical that it be straightforward for a user to
473 cut and paste an exact copy of the NSID payload output by a debugging
474 tool into other formats such as email messages or web forms without
475 distortion. Hexadecimal strings, while ugly, are also robust.
479 In some cases, adding the NSID option to a response message may
480 trigger message truncation. This specification does not change the
481 rules for DNS message truncation in any way, but implementors will
482 need to pay attention to this issue.
484 Including the NSID option in a response is always optional, so this
485 specification never requires name servers to truncate response
488 By definition, a resolver that requests NSID responses also supports
489 EDNS, so a resolver that requests NSID responses can also use the
490 "sender's UDP payload size" field of the OPT pseudo-RR to signal a
491 receive buffer size large enough to make truncation unlikely.
503 Austein Expires July 15, 2006 [Page 9]
505 Internet-Draft DNS NSID January 2006
508 4. IANA Considerations
510 This mechanism requires allocation of one ENDS option code for the
511 NSID option (Section 2.3).
559 Austein Expires July 15, 2006 [Page 10]
561 Internet-Draft DNS NSID January 2006
564 5. Security Considerations
566 This document describes a channel signaling mechanism, intended
567 primarily for debugging. Channel signaling mechanisms are outside
568 the scope of DNSSEC per se. Applications that require integrity
569 protection for the data being signaled will need to use a channel
570 security mechanism such as TSIG [RFC2845].
572 Section 3.1 discusses a number of different kinds of information that
573 a name server operator might choose to provide as the value of the
574 NSID option. Some of these kinds of information are security
575 sensitive in some environments. This specification deliberately
576 leaves the syntax and semantics of the NSID option content up to the
577 implementation and the name server operator.
615 Austein Expires July 15, 2006 [Page 11]
617 Internet-Draft DNS NSID January 2006
622 Joe Abley, Harald Alvestrand, Mark Andrews, Roy Arends, Steve
623 Bellovin, Randy Bush, David Conrad, Johan Ihren, Daniel Karrenberg,
624 Peter Koch, Mike Patton, Mike StJohns, Paul Vixie, Sam Weiler, and
625 Suzanne Woolf. Apologies to anyone inadvertently omitted from the
671 Austein Expires July 15, 2006 [Page 12]
673 Internet-Draft DNS NSID January 2006
678 7.1. Normative References
680 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
681 Requirement Levels", RFC 2119, BCP 14, March 1997.
683 [RFC2671] Vixie, P., "Extension Mechanisms for DNS (EDNS0)",
684 RFC 2671, August 1999.
686 [RFC2845] Vixie, P., Gudmundsson, O., Eastlake 3rd, D., and B.
687 Wellington, "Secret Key Transaction Authentication for DNS
688 (TSIG)", RFC 2845, May 2000.
690 7.2. Informative References
692 [RFC2277] Alvestrand, H., "IETF Policy on Character Sets and
693 Languages", RFC 2277, BCP 18, January 1998.
727 Austein Expires July 15, 2006 [Page 13]
729 Internet-Draft DNS NSID January 2006
737 Redwood City, CA 94063
783 Austein Expires July 15, 2006 [Page 14]
785 Internet-Draft DNS NSID January 2006
788 Intellectual Property Statement
790 The IETF takes no position regarding the validity or scope of any
791 Intellectual Property Rights or other rights that might be claimed to
792 pertain to the implementation or use of the technology described in
793 this document or the extent to which any license under such rights
794 might or might not be available; nor does it represent that it has
795 made any independent effort to identify any such rights. Information
796 on the procedures with respect to rights in RFC documents can be
797 found in BCP 78 and BCP 79.
799 Copies of IPR disclosures made to the IETF Secretariat and any
800 assurances of licenses to be made available, or the result of an
801 attempt made to obtain a general license or permission for the use of
802 such proprietary rights by implementers or users of this
803 specification can be obtained from the IETF on-line IPR repository at
804 http://www.ietf.org/ipr.
806 The IETF invites any interested party to bring to its attention any
807 copyrights, patents or patent applications, or other proprietary
808 rights that may cover technology that may be required to implement
809 this standard. Please address the information to the IETF at
813 Disclaimer of Validity
815 This document and the information contained herein are provided on an
816 "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS
817 OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET
818 ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED,
819 INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE
820 INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
821 WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
826 Copyright (C) The Internet Society (2006). This document is subject
827 to the rights, licenses and restrictions contained in BCP 78, and
828 except as set forth therein, the authors retain all their rights.
833 Funding for the RFC Editor function is currently provided by the
839 Austein Expires July 15, 2006 [Page 15]