idnits 2.17.1 draft-ietf-v6ops-renumbering-procedure-03.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** It looks like you're using RFC 3978 boilerplate. You should update this to the boilerplate described in the IETF Trust License Policy document (see https://trustee.ietf.org/license-info), which is required now. -- Found old boilerplate from RFC 3978, Section 5.1.a on line 17. -- Found old boilerplate from RFC 3978, Section 5.5 on line 946. -- Found old boilerplate from RFC 3979, Section 5, paragraph 1 on line 923. -- Found old boilerplate from RFC 3979, Section 5, paragraph 2 on line 930. -- Found old boilerplate from RFC 3979, Section 5, paragraph 3 on line 936. ** The document seems to lack an RFC 3978 Section 5.1 IPR Disclosure Acknowledgement. ** This document has an original RFC 3978 Section 5.4 Copyright Line, instead of the newer IETF Trust Copyright according to RFC 4748. ** This document has an original RFC 3978 Section 5.5 Disclaimer, instead of the newer disclaimer which includes the IETF Trust according to RFC 4748. ** The document uses RFC 3667 boilerplate or RFC 3978-like boilerplate instead of verbatim RFC 3978 boilerplate. After 6 May 2005, submission of drafts without verbatim RFC 3978 boilerplate is not accepted. The following non-3978 patterns matched text found in the document. That text should be removed or replaced: This document is an Internet-Draft and is subject to all provisions of Section 3 of RFC 3667. By submitting this Internet-Draft, each author represents that any applicable patent or other IPR claims of which he or she is aware have been or will be disclosed, and any of which he or she becomes aware will be disclosed, in accordance with Section 6 of BCP 79. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- == No 'Intended status' indicated for this document; assuming Proposed Standard Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** The document seems to lack an IANA Considerations section. (See Section 2.2 of https://www.ietf.org/id-info/checklist for how to handle the case when there are no actions for IANA.) Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the RFC 3978 Section 5.4 Copyright Line does not match the current year -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (November 22, 2004) is 7095 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) ** Downref: Normative reference to an Informational RFC: RFC 2072 ** Obsolete normative reference: RFC 2460 (Obsoleted by RFC 8200) ** Obsolete normative reference: RFC 2461 (Obsoleted by RFC 4861) ** Obsolete normative reference: RFC 2462 (Obsoleted by RFC 4862) ** Obsolete normative reference: RFC 3315 (Obsoleted by RFC 8415) == Outdated reference: A later version (-12) exists of draft-ietf-dnsop-ipv6-dns-issues-10 -- Obsolete informational reference (is this intentional?): RFC 1305 (Obsoleted by RFC 5905) -- Obsolete informational reference (is this intentional?): RFC 2535 (Obsoleted by RFC 4033, RFC 4034, RFC 4035) -- Obsolete informational reference (is this intentional?): RFC 2845 (Obsoleted by RFC 8945) -- Obsolete informational reference (is this intentional?): RFC 3177 (Obsoleted by RFC 6177) -- Obsolete informational reference (is this intentional?): RFC 3633 (Obsoleted by RFC 8415) Summary: 11 errors (**), 0 flaws (~~), 3 warnings (==), 12 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 1 IPv6 Operations Working Group F. Baker 2 Internet-Draft E. Lear 3 Expires: May 23, 2005 R. Droms 4 Cisco Systems 5 November 22, 2004 7 Procedures for Renumbering an IPv6 Network without a Flag Day 8 draft-ietf-v6ops-renumbering-procedure-03 10 Status of this Memo 12 This document is an Internet-Draft and is subject to all provisions 13 of section 3 of RFC 3667. By submitting this Internet-Draft, each 14 author represents that any applicable patent or other IPR claims of 15 which he or she is aware have been or will be disclosed, and any of 16 which he or she become aware will be disclosed, in accordance with 17 RFC 3668. 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 22 Internet-Drafts. 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 May 23, 2005. 37 Copyright Notice 39 Copyright (C) The Internet Society (2004). 41 Abstract 43 This document describes a procedure that can be used to renumber a 44 network from one prefix to another. It uses IPv6's intrinsic ability 45 to assign multiple addresses to a network interface to provide 46 continuity of network service through a "make-before-break" 47 transition, as well as addressing naming and configuration management 48 issues. It also uses other IPv6 features to minimize the effort and 49 time required to complete the transition from the old prefix to the 50 new prefix. 52 Table of Contents 54 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 55 1.1 Summary of the renumbering procedure . . . . . . . . . . . 3 56 1.2 Terminology . . . . . . . . . . . . . . . . . . . . . . . 4 57 1.3 Summary of what must be changed . . . . . . . . . . . . . 4 58 1.4 Multihoming Issues . . . . . . . . . . . . . . . . . . . . 5 60 2. Detailed review of procedure . . . . . . . . . . . . . . . . . 6 61 2.1 Initial condition: stable using the old prefix . . . . . . 6 62 2.2 Preparation for the renumbering process . . . . . . . . . 6 63 2.2.1 Domain Name Service . . . . . . . . . . . . . . . . . 7 64 2.2.2 Mechanisms for address assignment to interfaces . . . 8 65 2.3 Configuring switches and routers for the new prefix . . . 8 66 2.4 Adding new host addresses . . . . . . . . . . . . . . . . 9 67 2.5 Stable use of either prefix . . . . . . . . . . . . . . . 10 68 2.6 Transition from use of the old prefix to the new prefix . 10 69 2.6.1 Transition of DNS service to the new prefix . . . . . 10 70 2.6.2 Transition to the use of new addresses . . . . . . . . 10 71 2.7 Removing the old prefix . . . . . . . . . . . . . . . . . 11 72 2.8 Final condition: stable using the new prefix . . . . . . . 12 74 3. How to avoid shooting yourself in the foot . . . . . . . . . . 13 75 3.1 "Find all the places..." . . . . . . . . . . . . . . . . . 13 76 3.2 Renumbering switch and router interfaces . . . . . . . . . 13 77 3.3 Ingress Filtering . . . . . . . . . . . . . . . . . . . . 14 79 4. Call to Action for the IETF . . . . . . . . . . . . . . . . . 15 80 4.1 Dynamic updates to DNS across administrative domains . . . 15 81 4.2 Management of the reverse zone . . . . . . . . . . . . . . 15 83 5. Security Considerations . . . . . . . . . . . . . . . . . . . 16 85 6. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 18 87 7. References . . . . . . . . . . . . . . . . . . . . . . . . . . 19 88 7.1 Normative References . . . . . . . . . . . . . . . . . . . . 19 89 7.2 Informative References . . . . . . . . . . . . . . . . . . . 19 91 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . 20 93 A. Managing Latency in the DNS . . . . . . . . . . . . . . . . . 22 95 Intellectual Property and Copyright Statements . . . . . . . . 24 97 1. Introduction 99 The Prussian military theorist Carl von Clausewitz [Clausewitz] 100 wrote, "Everything is very simple in war, but the simplest thing is 101 difficult. These difficulties accumulate and produce a friction, 102 which no man can imagine exactly who has not seen war. ... So in 103 war, through the influence of an "infinity of petty circumstances" 104 which cannot properly be described on paper, things disappoint us and 105 we fall short of the mark." Operating a network is aptly compared to 106 conducting a war. The difference is that the opponent has the futile 107 expectation that homo ignoramus will behave intelligently. 109 A "flag day" is a procedure in which the network, or a part of it, is 110 changed during a planned outage, or suddenly, causing an outage while 111 the network recovers. Avoiding outages requires the network to be 112 modified using what in mobility might be called a "make before break" 113 procedure: the network is enabled to use a new prefix while the old 114 one is still operational, operation is switched to that prefix, and 115 then the old one is taken down. 117 This document addresses the key procedural issues in renumbering an 118 IPv6 [RFC2460] network without a "flag day". The procedure is 119 straightforward to describe, but operationally can be difficult to 120 automate or execute due to issues of statically configured network 121 state, which one might aptly describe as "an infinity of petty 122 circumstances". As a result, in certain areas, this procedure is 123 necessarily incomplete, as network environments vary widely and no 124 one solution fits all. It points out a few of many areas where there 125 are multiple approaches. It may be considered an update to RFC 2072 126 [RFC2072]. This document also contains recommendations for 127 application design and network management which, if taken seriously, 128 may avoid or minimize the impact of the issues. 130 1.1 Summary of the renumbering procedure 132 By "renumbering a network" we mean replacing the use of an existing 133 (or "old") prefix throughout a network with a new prefix. Usually, 134 both prefixes will be the same length. The procedures described in 135 this document are, for the most part, equally applicable if the two 136 prefixes are not the same length. During renumbering, sub-prefixes 137 (or "link prefixes") from the old prefix, which have been assigned to 138 links throughout the network, will be replaced by link prefixes from 139 the new prefix. Interfaces on systems throughout the network will be 140 configured with IPv6 addresses from the link prefixes of the new 141 prefix, and any addresses from the old prefix in services like DNS 142 [RFC1034][RFC1035] or configured into switches and routers and 143 applications will be replaced by the appropriate addresses from the 144 new prefix. 146 The renumbering procedure described in this document can be applied 147 to part of a network as well as an organization's entire network. In 148 the case of a large organization, it may be advantageous to treat the 149 network as a collection of smaller networks. Renumbering each of the 150 smaller networks separately will make the process more manageable. 151 The process described in this document is generally applicable to any 152 network, whether it is an entire organization network or part of a 153 larger network. 155 1.2 Terminology 157 DDNS: Dynamic DNS [RFC2136][RFC3007] updates can be secured through 158 the use of SIG(0)[RFC2535][RFC2931] and TSIG [RFC2845] 160 DHCP prefix delegation: An extension to DHCP [RFC3315] to automate 161 the assignment of a prefix; for example from an ISP to a 162 customer[RFC3633] 164 flag day: A transition which involves a planned service outage 166 ingress/egress filters: Filters applied to a router interface 167 connected to an external organization, such as an ISP, to exclude 168 traffic with inappropriate IPv6 addresses 170 link prefix: A prefix, usually a /64 [RFC3177], assigned to a link 172 SLAC: StateLess Address AutoConfiguration [RFC2462] 174 1.3 Summary of what must be changed 176 Addresses from the old prefix that are affected by renumbering will 177 appear in a wide variety of places in the components in the 178 renumbered network. The following list gives some of the places 179 which may include prefixes or addresses that are affected by 180 renumbering, and gives some guidance about how the work required 181 during renumbering might be minimized: 183 o Link prefixes assigned to links. Each link in the network must be 184 assigned a link prefix from the new prefix. 186 o IPv6 addresses assigned to interfaces on switches and routers. 187 These addresses are typically assigned manually, as part of 188 configuring switches and routers. 190 o Routing information propagated by switches and routers 192 o Link prefixes advertised by switches and routers. [RFC2461] 193 o Ingress/egress filters. 195 o ACLs and other embedded addresses on switches and routers. 197 o IPv6 addresses assigned to interfaces on hosts. Use of StateLess 198 Address Configuration [RFC2462] (SLAC) or DHCP [RFC3315] can 199 mitigate the impact of renumbering the interfaces on hosts. 201 o DNS entries. New AAAA and PTR records are added and old ones 202 removed in several phases to reflect the change of prefix. 203 Caching times are adjusted accordingly during these phases. 205 o IPv6 addresses and other configuration information provided by 206 DHCP. 208 o IPv6 addresses embedded in configuration files, applications and 209 elsewhere. Finding everything that must be updated and automating 210 the process may require significant effort, which is discussed in 211 more detail in Section 3. This process must be tailored to the 212 needs of each network. 214 1.4 Multihoming Issues 216 In addition to the considerations presented, the operational matters 217 of multihoming may need to be addressed. Networks are generally 218 renumbered for one of three reasons: the network itself is changing 219 its addressing policy and must renumber to implement the new policy 220 (for example, a company has been acquired and is changing addresses 221 to those used by its new owner), an upstream provider has changed its 222 prefixes and its customers are forced to do so at the same time, or a 223 company is changing providers and must perforce use addresses 224 assigned by the new provider. The third case is common. 226 When a company changes providers, it is common to institute an 227 overlap period, during which it is served by both providers. By 228 definition, the company is multihomed during such a period. While 229 this document is not about multihoming per se, problems can arise as 230 a result of ingress filtering policies applied by the upstream 231 provider or one of its upstream providers, so the user of this 232 document need also be cognizant of these issues. This is discussed 233 in detail, and approaches to dealing with it are described, in 234 [RFC2827] and [RFC3704]. 236 2. Detailed review of procedure 238 During the renumbering process, the network transitions through eight 239 states. In the initial state, the network uses just the prefix which 240 is to be replaced during the renumbering process. At the end of the 241 process, the old prefix has been entirely replaced by the new prefix, 242 and the network is using just the new prefix. To avoid a flag day 243 transition, the new prefix is deployed first and the network reaches 244 an intermediate state in which either prefix can be used. In this 245 state, individual hosts can make the transition to using the new 246 prefix as appropriate to avoid disruption of applications. Once all 247 of the hosts have made the transition to the new prefix, the network 248 is reconfigured so that the old prefix is no longer used in the 249 network. 251 In this discussion, we assume that an entire prefix is being replaced 252 with another entire prefix. It may be that only part of a prefix is 253 being changed, or that more than one prefix is being changed to a 254 single joined prefix. In such cases, the basic principles apply, but 255 will need to be modified to address the exact situation. This 256 procedure should be seen as a skeleton of a more detailed procedure 257 that has been tailored to a specific environment. Put simply, season 258 to taste. 260 2.1 Initial condition: stable using the old prefix 262 Initially, the network is using an old prefix in routing, device 263 interface addresses, filtering, firewalls and other systems. This is 264 a stable configuration. 266 2.2 Preparation for the renumbering process 268 The first step is to obtain the new prefix and new reverse zone from 269 the delegating authority. These delegations are performed using 270 established procedures, from either an internal or external 271 delegating authority. 273 Before any devices are reconfigured as a result of the renumbering 274 event, each link in the network must be assigned a sub-prefix from 275 the new prefix. While this assigned link prefix doesn't explicitly 276 appear in the configuration of any specific switch, router, or host, 277 the network administrator performing the renumbering procedure must 278 make these link prefix assignments prior to beginning the procedure 279 to guide the configuration of switches and routers, assignment of 280 addresses to interfaces and modifications to network services such as 281 DNS and DHCP. 283 Prior to renumbering, various processes will need to be reconfigured 284 to confirm bindings between names and addresses more frequently. In 285 normal operation, DNS name translations and DHCP bindings are often 286 given relatively long lifetimes to limit server load. In order to 287 reduce transition time from old to new prefix it may be necessary to 288 reduce the time to live (TTL) associated with DNS records and 289 increase the frequency with which DHCP clients contact the DHCP 290 server. At the same time, a procedure must be developed through 291 which other configuration parameters will be updated during the 292 transition period when both prefixes are available. 294 2.2.1 Domain Name Service 296 During the renumbering process, the DNS database must be updated to 297 add information about addresses assigned to interfaces from the new 298 prefix and to remove addresses assigned to interfaces from the old 299 prefix. The changes to the DNS must be coordinated with the changes 300 to the addresses assigned to interfaces. 302 Changes to the information in the DNS have to propagate from the 303 server at which the change was made to the resolvers where the 304 information is used. The speed of this propagation is controlled by 305 the TTL for DNS records and the frequency of updates from primary to 306 secondary servers. 308 The latency in propagating changes in the DNS can be managed through 309 the TTL assigned to individual DNS records and through the timing of 310 updates from primary to secondary servers. Appendix A gives an 311 analysis of the factors controlling the propagation delays in the 312 DNS. 314 The suggestions for reducing the delay in the transition to new IPv6 315 addresses applies when the DNS service can be given prior notice 316 about a renumbering event. However, the DNS service for a host may 317 be in a different administrative domain than the network to which the 318 host is attached. For example, a device from organization A that 319 roams to a network belonging to organization B, the device's DNS A 320 record is still managed by organization A, where the DNS service 321 won't be given advance notice of a renumbering event in organization 322 B. 324 One strategy for updating the DNS is to allow each system to manage 325 its own DNS information through Dynamic DNS (DDNS) 326 [RFC2136][RFC3007]. Authentication of these DDNS updates is strongly 327 recommended, and can be accomplished through TSIG and SIG(0). Both 328 TSIG and SIG(0) require configuration and distribution of keys to 329 hosts and name servers in advance of the renumbering event. 331 2.2.2 Mechanisms for address assignment to interfaces 333 IPv6 addresses may be assigned through SLAC, DHCP, and manual 334 processes. If DHCP is used for IPv6 address assignment, there may be 335 some delay in the assignment of IPv6 addresses from the new prefix 336 because hosts using DHCP only contact the server periodically to 337 extend the lifetimes on assigned addresses. This delay can be 338 reduced in two ways: 340 o Prior to the renumbering event, the T1 parameter (which controls 341 the time at which a host using DHCP contacts the server) may be 342 reduced. 344 o The DHCP Reconfigure message may also be sent from the server to 345 the hosts to trigger the hosts to contact the server immediately. 347 2.3 Configuring switches and routers for the new prefix 349 In this step, switches and routers and services are prepared for the 350 new prefix but the new prefix is not used for any datagram 351 forwarding. Throughout this step, the new prefix is added to the 352 network infrastructure in parallel with (and without interfering 353 with) the old prefix. For example, addresses assigned from the new 354 prefix are configured in addition to any addresses from the old 355 prefix assigned to interfaces on the switches and routers. Changes 356 to the routing infrastructure for the new prefix are added in 357 parallel with the old prefix so that forwarding for both prefixes 358 operates in parallel. At the end of this step, the network is still 359 running on the old prefix but is ready to begin using the new prefix. 361 The new prefix is added to the routing infrastructure, firewall 362 filters, ingress/egress filters and other forwarding and filtering 363 functions. Routes for the new link prefixes may be injected by 364 routing protocols into the routing subsystem, but the router 365 advertisements should not cause hosts to perform SLAC on the new link 366 prefixes; in particular the "autonomous address-configuration" flag 367 [RFC2461] should not be set in the advertisements for the new link 368 prefixes. The reason hosts should not be forming addresses at this 369 point is that routing to the new addresses may not yet be stable. 371 The details of this step will depend on the specific architecture of 372 the network being renumbered and the capabilities of the components 373 that make up the network infrastructure. The effort required to 374 complete this step may be mitigated by the use of DNS, DHCP prefix 375 delegation [RFC3633] and other automated configuration tools. 377 While the new prefix is being added, it will of necessity not be 378 working everywhere in the network, and unless properly protected by 379 some means such as ingress and egress access lists, the network may 380 be attacked through the new prefix in those places where it is 381 operational. 383 Once the new prefix has been added to the network infrastructure, 384 access-lists, route-maps and other network configuration options that 385 use IP addresses should be checked to ensure that hosts and services 386 that use the new prefix will behave as they did with the old one. 387 Name services other than DNS and other services that provide 388 information that will be affected by renumbering must be updated in 389 such a way as to avoid responding with stale information. There are 390 several useful approaches to identify and augment configurations: 392 Develop a mapping from each network and address derived from the 393 old prefix to each network and address derived from the new 394 prefix. Tools such as the UNIX "sed" or "perl" utilities are 395 useful to then find and augment access-lists, route-maps, and the 396 like. 398 A similar approach involves the use of such mechanisms as DHCP 399 prefix delegation to abstract networks and addresses. 401 Switches and routers or manually configured hosts that have IPv6 402 addresses assigned from the new prefix may be used at this point to 403 test the network infrastructure. 405 Advertisement of the prefix outside its network is the last thing to 406 be configured during this phase. One wants to have all of one's 407 defenses in place before advertising the prefix, if only because the 408 prefix may come under immediate attack. 410 At the end of this phase routing, access control, and other network 411 services should work interchangeably for both old and new prefixes. 413 2.4 Adding new host addresses 415 Once the network infrastructure for the new prefix are in place and 416 tested, IPv6 addresses from the new prefix may be assigned to host 417 interfaces. These IPv6 addresses may be assigned through SLAC, DHCP, 418 and manual processes. If SLAC is used in the network, the switches 419 and routers are configured to indicate that hosts should use SLAC to 420 assign IPv6 addresses from the new prefix. If DHCP is used for IPv6 421 address assignment, the DHCP service is configured to assign IPv6 422 addresses to hosts. 424 Once the new IPv6 addresses have been assigned to the host 425 interfaces, both the forward and reverse maps within DNS should be 426 updated for the new addresses, either through automated or manual 427 means. In particular, some clients may be able to update their 428 forward maps through DDNS, while automating the update of the reverse 429 zone may be more difficult as discussed in Section 4.2. 431 2.5 Stable use of either prefix 433 Once the network has been configured with the new prefix and has had 434 sufficient time to stabilize, it becomes a stable platform with two 435 addresses configured on each and every infrastructure component 436 interface (apart from interfaces that use only the link-local 437 address), and two non-link-local addresses are available for the use 438 of any host, one in the old prefix and one in the new. This is a 439 stable configuration. 441 2.6 Transition from use of the old prefix to the new prefix 443 When the new prefix has been fully integrated into the network 444 infrastructure and has been tested for stable operation, hosts and 445 switches and routers can begin using the new prefix. Once the 446 transition has completed the old prefix will not be in use in the 447 network. 449 2.6.1 Transition of DNS service to the new prefix 451 The DNS service is configured to use the new prefix by removing any 452 IPv6 addresses from the old prefix from the DNS server configuration. 453 External references to the DNS servers, such as in the DNS service 454 from which this DNS domain was delegated, are updated to use the IPv6 455 addresses from the new prefix. 457 2.6.2 Transition to the use of new addresses 459 When both prefixes are usable in the network, each host can make the 460 transition from using the old prefix to the new prefix at a time that 461 is appropriate for the applications on the host. If the host 462 transitions are randomized, DNS dynamic update mechanisms can better 463 scale to accommodate the changes to the DNS. 465 As services become available through addresses from the new prefix, 466 references to the hosts providing those services are updated to use 467 the new prefix. Addresses obtained through DNS will be automatically 468 updated when the DNS names are resolved. Addresses may also be 469 obtained through DHCP, and will be updated as hosts contact DHCP 470 servers. Addresses that are otherwise configured must be updated 471 appropriately. 473 It may be necessary to provide users with tools or other explicit 474 procedures to complete the transition from the use of the old prefix 475 to the new prefix, because some applications and operating system 476 functions may be configured in ways that do not use DNS at all or 477 will not use DNS to resolve a domain name to a new address once the 478 new prefix is available. For example, a device that only uses DNS to 479 resolve the name of an NTP server when the device is initialized will 480 not obtain the address from the new prefix for that server at this 481 point in the renumbering process. 483 This last point warrants repeating (in a slightly different form). 484 Applications may cache addressing information in different ways, for 485 varying lengths of time. They may cache this information in memory, 486 on a file system, or in a database. Only after careful observation 487 and consideration of one"s environment should one conclude that a 488 prefix is no longer in use. For more information on this issue, 489 please see [I-D.ietf-dnsop-ipv6-dns-issues]. 491 The transition of critical services, such as DNS, DHCP, NTP [RFC1305] 492 and important business services should be managed and tested 493 carefully to avoid service outages. Each host should take reasonable 494 precautions prior to changing to the use of the new prefix to 495 minimize the chance of broken connections. For example, utilities 496 such as netstat and network analyzers can be used to determine if any 497 existing connections to the host are still using the address from the 498 old prefix for that host. 500 Link prefixes from the old prefix in router advertisements and 501 addresses from the old prefix provided through DHCP should have their 502 preferred lifetimes set to zero at this point, so that hosts will not 503 use the old prefixes for new communications. 505 2.7 Removing the old prefix 507 Once all sessions are deemed to have completed, there will be no 508 dependence on the old prefix. It may be removed from the 509 configuration of the routing system, and from any static 510 configurations that depend on it. If any configuration has been 511 created based on DNS information, the configuration should be 512 refreshed after the old prefixes have been removed from the DNS. 514 During this phase the old prefix may be reclaimed by the provider or 515 Regional Internet Registry that granted it, and addresses within that 516 prefix are removed DNS. 518 In addition, DNS reverse maps for the old prefix may be removed from 519 the primary name server and the zone delegation may be removed from 520 the parent zone. Any DNS, DHCP, or SLAC timers that were changed 521 should be reset to their original values (most notably the DNS 522 forward map TTL). 524 2.8 Final condition: stable using the new prefix 526 This is equivalent to the first state, but using the new prefix. 528 3. How to avoid shooting yourself in the foot 530 The difficult operational issues in Section 2.3, Section 2.6, and 531 Section 2.7 are in dealing with the configurations of routers and 532 hosts which are not under the control of the network administrator or 533 are manually configured. Examples of such devices include voice over 534 IP (VoIP) telephones with static configuration of boot or name 535 servers, dedicated devices used in manufacturing that are configured 536 with the IP addresses for specific services, the boot servers of 537 routers and switches, etc. 539 3.1 "Find all the places..." 541 Long-lived communications present a problem to any application, 542 regardless of whether a network is renumbered. However, such events 543 test the resilience of the code used to survive disruptions. Many 544 existing applications make use of standard POSIX functions such as 545 gethostbyname() and getipnodebyname(), both of which use the common 546 "hostent" structure. If the application caches the response or for 547 whatever reason actually records the response on disk, recovery from 548 such events may prove difficult. It should be pointed out that the 549 above basic functions do not preserve DNS caching semantics. Any 550 application that requires repeated use of an IP address should either 551 not cache the result or make use of an appropriate function such as 552 the POSIX res_query(). 554 In some cases, applications make use of manually configured IP 555 addresses. For instance, database servers and web servers are often 556 configured to LISTEN to a specific TCP port on a specific interface. 557 Applications should be configured with names that are translated at 558 startup, and then again when the interface address itself is changed. 560 In the case of infrastructure applications such as name servers and 561 route processes, a convenient programmatic method that abstracts 562 prefixes to a single point would be ideal. 564 Application designers, equipment vendors, and the Open Source 565 community should take note. There is an opportunity to serve their 566 customers well in this area, and network operators should take note 567 to either develop or purchase appropriate tools. 569 3.2 Renumbering switch and router interfaces 571 The configuration and operation of switches and routers are often 572 designed to use static configuration with IP addresses or to resolve 573 domain names only once and use the resulting IP addresses until the 574 element is restarted. These static configurations complicate the 575 process of renumbering, requiring administration of all of the static 576 information and manual configuration during a renumbering event. 578 Because switches and routers are usually single-purpose devices, the 579 user interface and operating functions (software and hardware) are 580 often better integrated than independent services running on a server 581 platform. Thus, it is likely that switch vendors and router vendors 582 can design and implement consistent support for renumbering across 583 all of the functions of switches and routers. 585 To better support renumbering, switches and routers should use domain 586 names for configuration wherever appropriate, and should resolve 587 those names using the DNS when the lifetime on the name expires. 589 3.3 Ingress Filtering 591 An important consideration in Section 2.3, in the case where the 592 network being renumbered is connected to an external provider, the 593 network's ingress filtering policy and its provider's ingress 594 filtering policy. Both the network firewall's ingress filter and the 595 provider's ingress filter on the access link to the network should be 596 configured to prevent attacks that use source address spoofing. 597 Ingress filtering is considered in detail in "Ingress Filtering for 598 Multihomed Networks" [RFC3704]. 600 4. Call to Action for the IETF 602 The more automated one can make the renumbering process, the better 603 for everyone. Sadly, there are several mechanisms that either have 604 not been automated, or have not been automated consistently across 605 platforms. 607 4.1 Dynamic updates to DNS across administrative domains 609 The configuration files for a DNS server (such as named.conf) will 610 contain addresses that must be reconfigured manually during a 611 renumbering event. There is currently no easy way to automate the 612 update of these addresses, as the updates require both complex trust 613 relationships and automation to verify them. For instance, a reverse 614 zone is delegated by an upstream ISP, but there is currently no 615 mechanism to note additional delegations. 617 4.2 Management of the reverse zone 619 In networks where hosts obtain IPv6 addresses through SLAC, updates 620 of reverse zone are problematic because of lack of trust relationship 621 between administrative domain owning the prefix and the host 622 assigning the low 64 bits using SLAC. For example, suppose a host, 623 H, from organization A is connected to a network owned by 624 organization B. When H obtains a new address during a renumbering 625 event through SLAC, H will need to update its reverse entry in the 626 DNS through a DNS server from B that owns the reverse zone for the 627 new address. For H to update its reverse entry, the DNS server from 628 B must accept a DDNS request from H, requiring that an 629 inter-administrative domain trust relationship exist between H and B. 630 The IETF should develop a BCP recommendation for addressing this 631 problem. 633 5. Security Considerations 635 The process of renumbering is straightforward in theory but can be 636 difficult and dangerous in practice. The threats fall into two broad 637 categories: those arising from misconfiguration and those which are 638 actual attacks. 640 Misconfigurations can easily arise if any system in the network 641 "knows" the old prefix, or an address in it, a priori and is not 642 configured with the new prefix, or if the new prefix is configured in 643 a manner which replaces the old instead of being co-equal to it for a 644 period of time. Simplistic examples include: 646 Neglecting to reconfigure a system that is using the old prefix in 647 some static configuration: In this case, when the old prefix is 648 removed from the network, whatever feature was so configured 649 becomes inoperative - it is not configured for the new prefix, and 650 the old prefix is irrelevant. 652 Configuring a system via an IPv6 address, and replacing that old 653 address with a new address: Because the TCP connection is using the 654 old and now invalid IPv6 address, the SSH session will be 655 terminated and you will have to use SSH through the new address 656 for additional configuration changes. 658 Removing the old configuration before supplying the new: In this 659 case, it may be necessary to obtain on-site support or travel to 660 the system and access it via its console. 662 Clearly, taking the extra time to add the new prefix to the 663 configuration, allow the network to settle, and then remove the old 664 obviates this class of issue. A special consideration applies when 665 some devices are only occasionally used; the administration must 666 allow sufficiently long in Section 2.6 to ensure that their 667 likelihood of detection is sufficiently high. 669 A subtle case of this type can result when the DNS is used to 670 populate access control lists and similar security or QoS 671 configurations. DNS names used to translate between system or 672 service names and corresponding addresses are treated in this 673 procedure as providing the address in the preferred prefix, which is 674 either the old or the new prefix but not both. Such DNS names 675 provide a means in Section 2.6 to cause systems in the network to 676 stop using the old prefix to access servers or peers and cause them 677 to start using the new prefix. DNS names used for access control 678 lists, however, need to go through the same three step procedure used 679 for other access control lists, having the new prefix added to them 680 in Section 2.3 and the old prefix removed in Section 2.7. 682 It should be noted that the use of DNS names in this way is not 683 universally accepted as a solution to this problem; [RFC3871] 684 especially notes cases where static IP addresses are preferred over 685 DNS names, in order to avoid a name lookup when the naming system is 686 inaccessible or when the result of the lookup may be one of several 687 interfaces or systems. In such cases, extra care must be taken to 688 manage renumbering properly. 690 Attacks are also possible. Suppose, for example, that the new prefix 691 has been presented by a service provider, and the service provider 692 starts advertising the prefix before the customer network is ready. 693 The new prefix might be targeted in a distributed denial of service 694 attack, or a system might be broken into using an application that 695 would not cross the firewall using the old prefix, before the 696 network's defenses have been configured. Clearly, one wants to 697 configure the defenses first and only then accessibility and routing, 698 as described in Section 2.3 and Section 3.3. 700 The SLAC procedure described in [RFC2462] renumbers hosts. Dynamic 701 DNS provides a capability for updating DNS accordingly. Managing 702 configuration items apart from those procedures is most obviously 703 straightforward if all such configurations are generated from a 704 central configuration repository or database, or if they can all be 705 read into a temporary database, changed using appropriate scripts, 706 and applied to the appropriate systems. Any place where scripted 707 configuration management is not possible or is not used must be 708 tracked and managed manually. Here, there be dragons. 710 In ingress filtering of a multihomed network, an easy solution to the 711 issues raised in Section 3.3 might recommend that ingress filtering 712 should not be done for multihomed customers or that ingress filtering 713 should be special-cased. However, this has an impact on Internet 714 security. A sufficient level of ingress filtering is needed to 715 prevent attacks using spoofed source addresses. Another problem 716 becomes from the fact that if ingress filtering is made too difficult 717 (e.g. by requiring special casing in every ISP doing it), it might 718 not be done at an ISP at all. Therefore, any mechanism depending on 719 relaxing ingress filtering checks should be dealt with an extreme 720 care. 722 6. Acknowledgments 724 This document grew out of a discussion on the IETF list. Commentary 725 on the document came from Bill Fenner, Christian Huitema, Craig 726 Huegen, Dan Wing. Fred Templin, Hans Kruse, Harald Tveit Alvestrand, 727 Iljitsch van Beijnum, Jeff Wells, John Schnizlein, Laurent Nicolas, 728 Michael Thomas, Michel Py, Ole Troan, Pekka Savola, Peter Elford, 729 Roland Dobbins, Scott Bradner, Sean Convery, and Tony Hain. 731 Some took it on themselves to convince the authors that the concept 732 of network renumbering as a normal or frequent procedure is daft. 733 Their comments, if they result in improved address management 734 practices in networks, may be the best contribution this note has to 735 offer. 737 Christian Huitema, Pekka Savola, and Iljitsch van Beijnum described 738 the ingress filtering issues. These made their way separately into 739 [RFC3704], which should be read and understood by anyone that will 740 temporarily or permanently create a multihomed network by renumbering 741 from one provider to another. 743 7. References 745 7.1 Normative References 747 [RFC1034] Mockapetris, P., "Domain names - concepts and facilities", 748 STD 13, RFC 1034, November 1987. 750 [RFC1035] Mockapetris, P., "Domain names - implementation and 751 specification", STD 13, RFC 1035, November 1987. 753 [RFC2072] Berkowitz, H., "Router Renumbering Guide", RFC 2072, 754 January 1997. 756 [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6 757 (IPv6) Specification", RFC 2460, December 1998. 759 [RFC2461] Narten, T., Nordmark, E. and W. Simpson, "Neighbor 760 Discovery for IP Version 6 (IPv6)", RFC 2461, December 761 1998. 763 [RFC2462] Thomson, S. and T. Narten, "IPv6 Stateless Address 764 Autoconfiguration", RFC 2462, December 1998. 766 [RFC3315] Droms, R., Bound, J., Volz, B., Lemon, T., Perkins, C. and 767 M. Carney, "Dynamic Host Configuration Protocol for IPv6 768 (DHCPv6)", RFC 3315, July 2003. 770 [RFC3704] Baker, F. and P. Savola, "Ingress Filtering for Multihomed 771 Networks", BCP 84, RFC 3704, March 2004. 773 7.2 Informative References 775 [Clausewitz] 776 von Clausewitz, C., Howard, M., Paret, P. and D. Brodie, 777 "On War, Chapter VII, 'Friction in War'", June 1989. 779 [I-D.ietf-dnsop-ipv6-dns-issues] 780 Durand, A., Ihren, J. and P. Savola, "Operational 781 Considerations and Issues with IPv6 DNS", 782 draft-ietf-dnsop-ipv6-dns-issues-10 (work in progress), 783 October 2004. 785 [RFC1305] Mills, D., "Network Time Protocol (Version 3) 786 Specification, Implementation", RFC 1305, March 1992. 788 [RFC1995] Ohta, M., "Incremental Zone Transfer in DNS", RFC 1995, 789 August 1996. 791 [RFC1996] Vixie, P., "A Mechanism for Prompt Notification of Zone 792 Changes (DNS NOTIFY)", RFC 1996, August 1996. 794 [RFC2136] Vixie, P., Thomson, S., Rekhter, Y. and J. Bound, "Dynamic 795 Updates in the Domain Name System (DNS UPDATE)", RFC 2136, 796 April 1997. 798 [RFC2535] Eastlake, D., "Domain Name System Security Extensions", 799 RFC 2535, March 1999. 801 [RFC2827] Ferguson, P. and D. Senie, "Network Ingress Filtering: 802 Defeating Denial of Service Attacks which employ IP Source 803 Address Spoofing", BCP 38, RFC 2827, May 2000. 805 [RFC2845] Vixie, P., Gudmundsson, O., Eastlake, D. and B. 806 Wellington, "Secret Key Transaction Authentication for DNS 807 (TSIG)", RFC 2845, May 2000. 809 [RFC2931] Eastlake, D., "DNS Request and Transaction Signatures ( 810 SIG(0)s)", RFC 2931, September 2000. 812 [RFC3007] Wellington, B., "Secure Domain Name System (DNS) Dynamic 813 Update", RFC 3007, November 2000. 815 [RFC3177] IAB and IESG, "IAB/IESG Recommendations on IPv6 Address 816 Allocations to Sites", RFC 3177, September 2001. 818 [RFC3633] Troan, O. and R. Droms, "IPv6 Prefix Options for Dynamic 819 Host Configuration Protocol (DHCP) version 6", RFC 3633, 820 December 2003. 822 [RFC3871] Jones, G., "Operational Security Requirements for Large 823 Internet Service Provider (ISP) IP Network 824 Infrastructure", RFC 3871, September 2004. 826 Authors' Addresses 828 Fred Baker 829 Cisco Systems 830 1121 Via Del Rey 831 Santa Barbara, CA 93117 832 US 834 Phone: 408-526-4257 835 Fax: 413-473-2403 836 EMail: fred@cisco.com 837 Eliot Lear 838 Cisco Systems 839 170 W. Tasman Dr. 840 San Jose, CA 95134 841 US 843 Phone: +1 408 527 4020 844 EMail: lear@cisco.com 846 Ralph Droms 847 Cisco Systems 848 200 Beaver Brook Road 849 Boxborough, MA 01719 850 US 852 Phone: +1 978 936-1674 853 EMail: rdroms@cisco.com 855 Appendix A. Managing Latency in the DNS 857 The procedure in this section can be used to determine and manage the 858 latency in updates to information a DNS resource record (RR). 860 There are several kinds of possible delays which are ignored in these 861 calculations: 863 o the time it takes for the administrators to make the changes, 865 o the time it may take to wait for the DNS update, if the 866 secondaries are only updated at regular intervals, and not 867 immediately, and 869 o the time the updating to all the secondaries takes. 871 Assume the use of NOTIFY [RFC1996] and IXFR [RFC1995] to transfer 872 updated information from the primary DNS server to any secondary 873 servers; this is a very quick update process, and the actual time to 874 update of information is not considered significant. 876 There's a target time, TC, at which we want to change the contents of 877 a DNS RR. The RR is currently configured with TTL == TTLOLD. Any 878 cached references to the RR will expire no more than TTLOLD in the 879 future. 881 At time TC - (TTLOLD + TTLNEW), the RR in the primary is configured 882 with TTLNEW (TTLNEW < TTLOLD). The update process is initiated to 883 push the RR to the secondaries. After the update, responses to 884 queries for the RR are returned with TTLNEW. There are still some 885 cached references with TTLOLD. 887 At time TC - TTLNEW, the RR in the primary is configured with the new 888 address. The update process is initiated to push the RR to the 889 secondaries. After the update, responses to queries for the RR 890 return the new address. All the cached references have TTLNEW. 891 Between this time and TC, responses to queries for the RR may be 892 returned with either the old address or the new address. This 893 ambiguity is acceptable, assuming the host is configured to respond 894 to both addresses. 896 At time TC all the cached references with the old address have 897 expired, and all subsequent queries will return the new address. 898 After TC (corresponding to the final state described in Section 2.8), 899 the TTL on the RR can be set to the initial value TTLOLD. 901 The network administrator can choose TTLOLD and TTLNEW to meet local 902 requirements. 904 As a concrete example, consider a case where TTLOLD is a week (168 905 hours), and TTLNEW is an hour. The preparation for the change of 906 addresses begins 169 hours before the address change. After 168 907 hours have passed and only one hour is left, the TTLNEW has 908 propagated everywhere, and one can change the address record(s). 909 These are propagated within the hour, after which one can restore TTL 910 value to a larger value. This approach minimizes time where it's 911 uncertain what kind of (address) information is returned from the 912 DNS. 914 Intellectual Property Statement 916 The IETF takes no position regarding the validity or scope of any 917 Intellectual Property Rights or other rights that might be claimed to 918 pertain to the implementation or use of the technology described in 919 this document or the extent to which any license under such rights 920 might or might not be available; nor does it represent that it has 921 made any independent effort to identify any such rights. Information 922 on the procedures with respect to rights in RFC documents can be 923 found in BCP 78 and BCP 79. 925 Copies of IPR disclosures made to the IETF Secretariat and any 926 assurances of licenses to be made available, or the result of an 927 attempt made to obtain a general license or permission for the use of 928 such proprietary rights by implementers or users of this 929 specification can be obtained from the IETF on-line IPR repository at 930 http://www.ietf.org/ipr. 932 The IETF invites any interested party to bring to its attention any 933 copyrights, patents or patent applications, or other proprietary 934 rights that may cover technology that may be required to implement 935 this standard. Please address the information to the IETF at 936 ietf-ipr@ietf.org. 938 Disclaimer of Validity 940 This document and the information contained herein are provided on an 941 "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS 942 OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET 943 ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, 944 INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE 945 INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED 946 WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. 948 Copyright Statement 950 Copyright (C) The Internet Society (2004). This document is subject 951 to the rights, licenses and restrictions contained in BCP 78, and 952 except as set forth therein, the authors retain all their rights. 954 Acknowledgment 956 Funding for the RFC Editor function is currently provided by the 957 Internet Society.