idnits 2.17.1 draft-briscoe-intarea-ipv4-id-reuse-04.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- -- The draft header indicates that this document updates RFC2003, but the abstract doesn't seem to mention this, which it should. -- The draft header indicates that this document updates RFC2780, but the abstract doesn't seem to mention this, which it should. -- The draft header indicates that this document updates RFC4301, but the abstract doesn't seem to mention this, which it should. -- The draft header indicates that this document updates RFC791, but the abstract doesn't seem to mention this, which it should. -- The draft header indicates that this document updates RFC4727, but the abstract doesn't seem to mention this, which it should. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == Line 850 has weird spacing: '...ificant bit: ...' (Using the creation date from RFC791, updated by this document, for RFC5378 checks: 1981-09-01) -- 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 (February 14, 2014) is 3717 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) == Unused Reference: 'Fransson04' is defined on line 976, but no explicit reference was found in the text Summary: 0 errors (**), 0 flaws (~~), 3 warnings (==), 7 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Internet Area Working Group B. Briscoe 3 Internet-Draft BT 4 Updates: 791, 2003, 2780, 4301, 4727, February 14, 2014 5 6864 (if approved) 6 Intended status: Standards Track 7 Expires: August 18, 2014 9 Reusing the IPv4 Identification Field in Atomic Packets 10 draft-briscoe-intarea-ipv4-id-reuse-04 12 Abstract 14 This specification takes a new approach to extensibility that is both 15 principled and a hack. It builds on recent moves to formalise the 16 increasingly common practice where fragmentation in IPv4 more closely 17 matches that of IPv6. The large majority of IPv4 packets are now 18 'atomic', meaning indivisible. In such packets, the 16 bits of the 19 IPv4 Identification (IPv4 ID) field are redundant and could be freed 20 up for the Internet community to put to other uses, at least within 21 the constraints imposed by their original use for reassembly. This 22 specification defines the process for redefining the semantics of 23 these bits. It uses the previously reserved control flag in the IPv4 24 header to indicate that these 16 bits have new semantics. Great care 25 is taken throughout to ease incremental deployment, even in the 26 presence of middleboxes that incorrectly discard or normalise packets 27 that have the reserved control flag set. 29 Status of This Memo 31 This Internet-Draft is submitted in full conformance with the 32 provisions of BCP 78 and BCP 79. 34 Internet-Drafts are working documents of the Internet Engineering 35 Task Force (IETF). Note that other groups may also distribute 36 working documents as Internet-Drafts. The list of current Internet- 37 Drafts is at http://datatracker.ietf.org/drafts/current/. 39 Internet-Drafts are draft documents valid for a maximum of six months 40 and may be updated, replaced, or obsoleted by other documents at any 41 time. It is inappropriate to use Internet-Drafts as reference 42 material or to cite them other than as "work in progress." 44 This Internet-Draft will expire on August 18, 2014. 46 Copyright Notice 48 Copyright (c) 2014 IETF Trust and the persons identified as the 49 document authors. All rights reserved. 51 This document is subject to BCP 78 and the IETF Trust's Legal 52 Provisions Relating to IETF Documents 53 (http://trustee.ietf.org/license-info) in effect on the date of 54 publication of this document. Please review these documents 55 carefully, as they describe your rights and restrictions with respect 56 to this document. Code Components extracted from this document must 57 include Simplified BSD License text as described in Section 4.e of 58 the Trust Legal Provisions and are provided without warranty as 59 described in the Simplified BSD License. 61 Table of Contents 63 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 64 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 5 65 3. IPv4 Wire Protocol Semantics for Reusing the 66 Identification Field . . . . . . . . . . . . . . . . . . . . . 5 67 4. Behaviour of Intermediate Nodes . . . . . . . . . . . . . . . 8 68 4.1. End-to-End Preservation of ID-Reuse Semantics . . . . . . 8 69 4.2. Tunnel Behaviour . . . . . . . . . . . . . . . . . . . . . 8 70 5. Process for Defining Subdivisions of the ID-Reuse Field . . . 9 71 5.1. Constraints on Uses of the ID-Reuse Field . . . . . . . . 10 72 5.2. Process Example . . . . . . . . . . . . . . . . . . . . . 11 73 6. Incremental Deployment of New Uses of the IPv4 ID Field . . . 13 74 6.2. Process for Using the ID-Reuse Field Without Requiring 75 RC=1 . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 76 7. Updates to Existing RFCs . . . . . . . . . . . . . . . . . . . 17 77 8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 18 78 9. Security Considerations . . . . . . . . . . . . . . . . . . . 19 79 10. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . 20 80 11. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 20 81 12. Outstanding Issues (to be removed when all resolved) . . . . . 21 82 13. References . . . . . . . . . . . . . . . . . . . . . . . . . . 21 83 13.1. Normative References . . . . . . . . . . . . . . . . . . . 21 84 13.2. Informative References . . . . . . . . . . . . . . . . . . 21 85 Appendix A. Why More Bits Cannot be Freed (To be Removed by 86 RFC Editor) . . . . . . . . . . . . . . . . . . . . . 22 87 Appendix B. Experimental or Standards Track? (To Be Removed 88 Before Publication) . . . . . . . . . . . . . . . . . 23 89 Appendix C. Changes in This Version (to be removed by RFC 90 Editor) . . . . . . . . . . . . . . . . . . . . . . . 24 92 Intended status: Standards Track? (to be removed before publication) 94 This draft defines a process and a protocol for enabling new 95 protocols, including their progression from experimental track to 96 standards track. A process specification cannot have lesser status 97 than the protocols it enables. So if this specification were to 98 start on the experimental track, it would not initially have 99 sufficient status to enable standards track protocols. 101 In order for the IETF to consider whether this draft itself should be 102 experimental or standards track, it has been written as if it is 103 intended for the standards track. Otherwise the parts of the process 104 for enabling standards track protocols would have had to have been 105 written hypothetically, which would have been highly confusing. If 106 the IETF decides this specification ought to start out on the 107 experimental track, the standards track parts of the process will 108 have to be edited out. 110 Appendix B discusses whether this draft itself would be better to 111 start as experimental or standards track. 113 1. Introduction 115 The Problem: The extensibility provisions in IP (v4 and v6) have 116 turned out not to be usable in practice. Hardware has been optimised 117 for the common case, so packets using extensibility mechanisms (e.g. 118 IPv4 options or IPv6 hop-by-hop options) are very likely to be punted 119 to the software slow-path and consequently likely to be dropped 120 whenever the software processor is busy [Fransson04, Cisco.IPv6Ext]. 122 This specification takes a different approach to extensibility. 123 Rather than flagging protocol extensions as 'extensions', it places 124 extension headers where they will be ignored by pre-existing 125 hardware. As code is added to routers to handle newly added 126 extensions, the code can tell the machine where to look for the 127 relevant header. 129 This approach recognises that extensions added after a protocol suite 130 was first defined are different to options defined as a coherent part 131 of the original protocol suite. Machines that have no code to 132 understand a protocol extension that was added later do not need to 133 punt a packet to the software processor merely to scan through chains 134 of headers that it will not know how to process. 136 Having only settled on this approach long after the TCP/IP suite has 137 been defined, it becomes necessary to find places in the existing 138 protocol headers that are already ignored by existing machines. In 139 an 'atomic' IPv4 packet, the Identification (IPv4 ID) field is one 140 such place that is redundant. This specification defines the process 141 through which the 16 bits in this field can be returned to the IETF 142 for use in future standards actions, at least within the constraints 143 imposed by their original use for reassembly. 145 Background: [RFC6864] updates IPv4 to more closely match the approach 146 taken to fragmentation in IPv6. It specifies that the IPv4 ID field 147 is only defined for 'atomic' packets. An atomic packet is one that 148 has not yet been fragmented (MF=0 and fragment offset=0) and for 149 which further fragmentation is inhibited (DF=1) [RFC6864]. 151 In practical scenarios, the IPv4 ID field is too small to guarantee 152 uniqueness during the lifetime of a packet anyway [RFC4963]. 153 Therefore it has become safer to disable fragmentation altogether and 154 instead use an approach such as packetization layer path MTU 155 discovery [RFC4821]. The large majority of IPv4 packets are now 156 atomic. 158 Approach: This specification defines the IPv4 control flag that was 159 previously reserved [RFC0791] as the Recycled flag (RC). An 160 implementation can set RC=1 in an atomic packet to unambiguously flag 161 that the IPv4 ID field is not to be interpreted as IP Identification, 162 but instead it has the alternative semantics of an ID-Reuse field. 163 By setting RC=1, IPv4 implementations can distinguish a value 164 deliberately written into the ID-Reuse field from the same value that 165 just happened to be written into the IP ID field of an atomic packet 166 by a pre-existing implementation. 168 Thus, this specification effectively uses up the last bit in the IPv4 169 header in order to free up 16 other bits. However, there are some 170 constraints on the use of these 16 bits due to their original use as 171 the IP ID field (enumerated in Section 5.1). Of course the main 172 constraint is that the bits are not available in non-atomic packets. 173 But fragmentation is now used only rarely anyway, so it makes sense 174 to see if the the Internet community can invent ways to use the 16 175 bits in the IPv4 ID field despite the constraints. 177 Frequently Asked Questions: 179 1. There are many cases where a non-compliant machine ignores Don't 180 Fragment (DF=1) and fragments a packet anyway. 182 One answer is that we cannot allow non-complaint behaviour to 183 always block progress. Another answer is that we may be able to 184 detect and circumvent such non-compliant behaviour. For 185 instance, if a non-compliant router fragments packets with DF=1, 186 it may be possible to enhance path maximum transmission unit 187 discovery (PMTUD) to find a lower segment size small enough to 188 prevent the offending box from fragmenting packets. 190 2. Shouldn't we be focusing on IPv6, not continuing to update IPv4? 192 The simple answer is that, where additions are made to IPv6, 193 sometimes it will be necessary to make a parallel addition to 194 IPv4, to ensure continued interoperability between the IETF's two 195 main protocols. 197 Document Roadmap: Section 3 defines the semantics of the updated IPv4 198 wire protocol and Section 4 defines intermediate node behaviour. 199 Section 5 defines the process to be used for reassigning sub-fields 200 of the IPv4 ID-Reuse field. Then Section 6 describes a way to 201 circumvent problems likely to arise when deploying this new protocol. 202 Finally, Section 7 enumerates the updates to pre-existing RFCs, 203 before the tailpiece sections considering IANA, Security and draw 204 conclusions. 206 2. Terminology 208 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 209 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 210 document are to be interpreted as described in RFC 2119 [RFC2119]. 212 Further terminology used within this document: 214 Atomic packet: A packet not yet having been fragmented (MF=0 and 215 fragment offset=0) and for which further fragmentation has been 216 inhibited (DF=1), or in the syntax of the C programming language 217 ((DF==1) && (MF==0) && (Offset==0)) [RFC6864]. 219 Recycled (RC) flag: The control flag that was 'reserved' in 220 [RFC0791] (Figure 1). The flag positioned at bit 48 of the IPv4 221 header (counting from 0). Alternatively, some would call this bit 222 0 (counting from 0) of octet 7 (counting from 1) of the IPv4 223 header. 225 ID-Reuse field: Octets 5 and 6 (counting from 1) of the IPv4 header 226 of an atomic packet (Figure 3). The field that would have been 227 the IP Identification field if the packet were not atomic. 229 3. IPv4 Wire Protocol Semantics for Reusing the Identification Field 231 This specification defines the control flag that was defined as 232 'reserved' in [RFC0791] as the Recycled (RC) flag (Figure 1). 234 0 1 2 235 +---+---+---+ 236 | R | D | M | 237 | C | F | F | 238 +---+---+---+ 240 The Recycled (RC) Flag was previously reserved. 242 Figure 1: The Control Flags at the Start of Byte 7 of the IPv4 Header 244 Figure 2 recaps the definitions of octets 5 to 8 (counting from 1) of 245 the IPv4 header [RFC0791]. 246 0 1 2 3 247 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 248 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 249 | Identification |Flags| Fragment Offset | 250 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 252 Figure 2: Recap of RFC791 Definition of Octets 5 to 8 of the IPv4 253 Header. 255 If an IPv4 implementation sets RC=1 on an atomic packet, octets 5 & 6 256 of the IPv4 header MUST be interpreted with the semantics of the ID- 257 Reuse field, and MUST NOT be interpreted as the Identification field. 258 Figure 3 shows how octets 5 & 6 are redefined as the ID-Reuse field 259 when the packet is atomic, in the case where RC=1. 260 0 1 2 3 261 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 262 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 263 | ID-Reuse |1 1 0|0 0 0 0 0 0 0 0 0 0 0 0 0| 264 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 266 The Identification Field is redefined as the ID-Reuse Field when the 267 Packet is Atomic and specifically when RC=1 269 Figure 3: Octets 5 to 8 of the IPv4 Header. 271 If the Recycled flag is cleared to RC=0 on an atomic packet, some 272 sub-fields of octets 5 & 6 of the IPv4 header MAY be interpreted with 273 the semantics of the ID-Reuse field, but only in the highly 274 constrained circumstances defined in Section 6.2. 276 For the avoidance of doubt, the Recycled flag alone MUST NOT be 277 assumed to indicate that the packet is atomic. Only the combination 278 of ((DF==1) && (MF==0) && (Offset==0)) indicates that a packet is 279 atomic. Then if the Recycled flag is also set, the ID field 280 unambiguously has the semantics of the ID-Reuse field. If the 281 Recycled flag of an atomic packet is cleared, its ID field only has 282 the semantics of the ID-Reuse field in specific limited 283 circumstances. 285 It is expected that proposals to use the ID-Reuse field will each 286 need a few bits, not the whole 16 bit field. Therefore this 287 specification establishes a new IANA registry (Section 8) to record 288 assignments of sub-divisions of the ID-Reuse field. In this way, it 289 will be possible for new uses of different sub-divisions to be 290 orthogonal to each other. The process for incrementally defining new 291 sub-divisions is specified in Section 5. 293 If an IPv4 packet header has RC=1 but it is not atomic ((DF==0) || 294 (MF==1) || (Offset !=0)), then all the fields of the IPv4 header are 295 undefined and reserved for future use. If an implementation receives 296 such a packet, it could imply: 298 o that some currently unknown attack is being attempted 300 o or that some future standards action has defined a meaning for 301 this reserved combination of header values 303 Therefore, if an implementation receives a non-atomic packets with 304 RC=1, it MUST treat the packet as if the Recycled flag were cleared 305 to 0, but it MUST NOT change the Recycled flag to zero. It MAY log 306 the arrival of such packets and/or raise an alarm. It MUST NOT 307 always drop such packets, but it MAY drop them under a policy that 308 can be revoked if it is established that the appearance of such 309 packets is the result of a future standards action. 311 For convenience only, the above rules are summarised in Table 1. The 312 semantics of octets 5 & 6 of the IPv4 header are tabulated for each 313 value of the RC flag (rows) and for whether the packet is atomic or 314 not (columns). 316 +---------+----------------+--------------------+ 317 | RC flag | Non-Atomic | Atomic | 318 +---------+----------------+--------------------+ 319 | 0 | Identification | ID-Reuse (Limited) | 320 | 1 | Undefined | ID-Reuse | 321 +---------+----------------+--------------------+ 323 Table 1: The Dependence of the Semantics of Octets 5 & 6 of the IPv4 324 Header on whether the Packet is Atomic and on the RC Flag 326 4. Behaviour of Intermediate Nodes 328 4.1. End-to-End Preservation of ID-Reuse Semantics 330 If the source sets the RC flag to 1 on an atomic packet, another node 331 MUST NOT clear the RC flag to zero. Otherwise the semantics of the 332 ID-Reuse field would change (see the Security Considerations in 333 Section 9 for discussion of the integrity of the ID-Reuse field). 334 Note that intermediate nodes are already not expected to change an 335 atomic packet to non-atomic, which otherwise would also risk changing 336 the semantics of the ID-Reuse field. 338 If the source zeros the RC flag on an atomic packet, an intermediate 339 node MAY change the RC flag to 1. At this time, no case is envisaged 340 where an intermediate node would need to do this. However, as this 341 behaviour preserves ID-Reuse semantics safely, it is not precluded in 342 case it will prove useful (e.g. for sender proxies). 344 4.2. Tunnel Behaviour 346 This specification does not need to change the following aspects of 347 IPv4-in-IPv4 tunnelling, which already provide the most useful 348 semantics for the ID-Reuse field: 350 o For some time, it has been mandated that an atomic packet "MUST" 351 be encapsulated by an atomic outer header [RFC2003] (although some 352 implementations are broken in this respect). 354 o On decapsulation the outgoing header will naturally propagate the 355 ID-Reuse field of the inner header. 357 However, compliant IPv4 encapsulation implementations SHOULD copy the 358 ID-Reuse field when encapsulating an atomic IPv4 packet in another 359 atomic IPv4 header, irrespective of the setting of the Recycled flag. 360 It would be ideal but impractical to assert 'MUST' in this last 361 clause, given it cannot be assumed that pre-existing IPv4-in-IPv4 362 encapsulators will propagate the ID-Reuse field to the outer header 363 (see Section 5.1). 365 IPv6 packets without a fragmentation extension header are inherently 366 atomic. Therefore, if an IPv4 header encapsulates an IPv6 packet, 367 the encapsulator is already required to set the outer as atomic. 369 There is no direct mapping between the IPv4 ID-Reuse field as a whole 370 and any IPv6 header field (main or extension), because the ID-Reuse 371 field is merely a container for yet-to-be-defined sub-fields. 372 However, sub-fields of the ID-Reuse field might be defined to provide 373 a mapping for IPv6 extension headers that need to be visible in the 374 outer IPv4 header of a tunnel. The present specification cannot say 375 anything in general about any such mappings or any associated tunnel 376 behaviour. Any such behaviour will have to be defined when 377 individual ID-Reuse sub-fields are specified. 379 5. Process for Defining Subdivisions of the ID-Reuse Field 381 When IPv4 was designed, then later IPv6, all the fields in the main 382 IP header were initially defined together in a coordinated fashion. 383 In contrast, the only practical way to define new uses for the bits 384 in the ID-Reuse field will be to adopt a gradual addition approach, 385 in which subsets of the bits or codepoints will have to be assigned 386 on the merits of each request at the time. 388 Each new scheme will need to submit an RFC that requests a 389 subdivision of the ID-Reuse field and assigns behaviours to the 390 codepoints within this subdivision. A specification defining a new 391 use of a subdivision of the ID-Reuse field MUST register this use 392 with the IANA, which will maintain a registry for this purpose 393 (Section 8). 395 Proposals to reuse the IP ID field could relate to other parts of the 396 IPv4 header in the following different ways {ToDo: this list is not 397 exhaustive}: 399 Orthogonal: Some new protocol proposals will need to apply whatever 400 is in the rest of the packet, e.g. whether unicast or multicast, 401 whatever the Diffserv codepoint and whatever else might have been 402 added in the rest of the IP-Reuse field. Schemes that need to be 403 orthogonal to other elements of the IPv4 protocol will require 404 assignment of a number of bits as a dedicated sub-field of the ID- 405 Reuse field. 407 Mutually exclusive: It might be impossible for two uses of the ID- 408 Reuse field to both apply to the same packet. Such mutually 409 exclusive schemes will only each require a range of codepoints 410 within a sub-field. 412 Conditional: Some protocol proposals might only apply when other 413 parts of the header satisfy certain conditions, e.g. only for 414 multicast packets. The IANA will need to register these 415 conditions so that the bits can still be assigned for other uses 416 when the conditions do not apply. 418 To allow interworking between sub-fields that are being defined 419 incrementally, every new protocol MUST assign the all-zeros codepoint 420 of its sub-field to mean the new protocol is 'turned off'. This 421 means that implementations of the new protocol will treat such 422 packets as they would have been treated before the new protocol was 423 defined. 425 Implementations MUST also clear to zero any bits in the ID-Reuse 426 field that are not defined at the time the implementation is coded. 428 Proposals to use sub-fields of ID-Reuse will have to be assessed in 429 the order they arrive without knowing what future proposals they 430 might preclude. To judge each proposal, at least the following 431 criteria will be used: 433 Constraint satisfaction: Each proposal MUST either satisfy all the 434 constraints in Section 5.1 below, or include measures to 435 circumvent them. 437 General usefulness: Proposals that are not applicable to a broad set 438 of services that can be built over the internetwork protocol 439 SHOULD NOT warrant consuming the newly freed up IPv4 header space. 441 Parsimony: Burning up a large proportion of the remaining bits will 442 count against a proposal. 444 Backward compatibility with prior uses of ID-Reuse: As more sub- 445 fields of the ID-Reuse field become defined, each new proposal 446 SHOULD ensure that it takes into account potential interactions 447 with earlier standards actions or experiments defining other sub- 448 fields. 450 Forward compatibility with potential uses of ID-Reuse: In addition, 451 proposals that demonstrate sensitivity to potential future uses of 452 the remaining sub-fields of the ID-Reuse field will be more likely 453 to progress through the IETF's approval process. 455 Do no harm: Proposals that do no harm to existing uses of the 456 Internet will be favoured over those that do more harm. 458 5.1. Constraints on Uses of the ID-Reuse Field 460 Atomic packets: The IPv4 ID field cannot be reused if the packet is 461 not atomic, because then the IP ID field will need to be used for 462 its original purpose: fragment reassembly. 464 IPsec interaction: The IP Authentication Header (AH) [RFC4302] 465 assumes and requires the IPv4 ID field to be immutable, otherwise 466 verification of authentication and integrity checks will fail. 467 Any new use of bits in the ID-Reuse field MUST ensure the bits are 468 immutable, at least between IPsec endpoints (whether transport or 469 tunnel mode). It cannot be assumed that pre-existing IPsec 470 implementations will check the setting of the Recycled flag. 472 Note that the Recycled flag itself is considered mutable and 473 masked out before calculating an authentication header [RFC4302] 474 (see Section 9). 476 Tunnelling: Any new use of the ID-Reuse field in atomic packets 477 cannot reliably assume that the ID-Reuse field will propagate 478 unchanged into the outer header of an IPv4-in-IPv4 tunnel 479 [RFC2003, RFC4301]. It is likely that an IPv4 tunnel ingress will 480 encapsulate an atomic packet with another atomic outer header, as 481 this behaviour was mandated in [RFC2003]. However it is known 482 that some implementations are broken in this respect. It is 483 possible that an IPv4 encapsulator might copy the IP ID field of 484 an arriving atomic packet into the outer header. However this 485 behaviour has never been required and therefore cannot be 486 guaranteed for pre-existing tunnels. 488 Nonetheless, it can be assumed that the IPv4 ID field will be 489 preserved through the inner header into the outgoing packet at the 490 other end of the tunnel (even though this behaviour would not 491 strictly have been necessary for an atomic packet). 493 Incremental deployment: Each new proposal will need to consider any 494 detrimental effects from pre-existing IPv4 implementations, 495 assuming that they are likely to act on atomic packets without 496 first checking on the setting of the Recycled flag. 498 5.2. Process Example 500 For illustration purposes, imagine two RFCs have been published: an 501 experimental track RFC called Experiment A (ExA) and a standards 502 track RFC called Standard B (StB) and . Imagine they define 503 respectively a use for bits bits 14 to 15 and 11 to 13 of the ID- 504 Reuse field. Figure 4 shows example IANA registry entries for these 505 imaginary sub-fields. 507 Protocol name: StB 508 RFC: BBBB 509 Leftmost bit: 11 510 No. of bits allocated: 3 511 Sub-field defined if: Atomic packet and RC=1 513 Protocol name: ExA 514 RFC: AAAA 515 Leftmost bit: 14 516 No. of bits allocated: 2 517 Sub-field defined if: Atomic packet and RC=1 519 Figure 4: Example IANA Registry of Sub-fields of the ID-Reuse Field 521 Figure 5 shows an example of how incremental specification of 522 subdivisions of ID-Reuse would work. 523 0 1 2 3 524 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 525 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 526 | ID-Reuse _____ ___|1 1 0|0 0 0 0 0 0 0 0 0 0 0 0 0| 527 |0 0 0 0 0 0 0 0 0 0 0| StB |ExA| | | 528 | |1 0 1|0 1| | | 529 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 531 Figure 5: Example of Reuse of Octets 5 & 6 using RC=1 533 The bits shown in each row of Figure 5 define the semantics of the 534 bits shown in the next row down, as follows: 536 o The top row identifies that the packet is atomic and the RC flag 537 is 1. Therefore octets 5 & 6 of the IPv4 header are redefined as 538 the ID-Reuse field. 540 o The middle row shows the bits assigned to Standard B and 541 Experiment A by IANA. An implementer has to ensure that all the 542 bits of the ID-Reuse field that are yet to be defined (bits 0-10) 543 are cleared to zero. 545 o The bottom row shows that an implementation of ExA has set its 546 2-bit sub-field to codepoint 01 and an implementation of StB has 547 set its 3-bit sub-field to codepoint 101. The meaning of each 548 would be defined in the RFCs for ExA and StB respectively. 550 Imagine now that Experiment C (ExC) is defined later to use bits 0-7 551 of the ID-Reuse field. If the packet in Figure 5 is received by an 552 implementation of ExC, then it will see only zeros in the ExC sub- 553 field. Therefore the implementation of ExC will treat the packet as 554 if ExC is turned off (as mandated in Section 5). 556 Similarly, the implementation of protocol StB can rely on being able 557 to turn off Experiment A by setting bits 14 & 15 to zero. 559 6. Incremental Deployment of New Uses of the IPv4 ID Field 561 When implementations first set the Recycled flag to 1, they are 562 likely to be blocked by certain middleboxes, either deliberately 563 (e.g. firewalls that assume anomalies are attacks) or erroneously 564 (e.g. having misunderstood the phrase "reserved, must be zero" in 565 RFC791). It is also possible that broken 'normalisers' might clear 566 RC to zero if it is 1, although so far no tests have found such 567 broken behaviour. 569 To address this problem, Section 6.2 introduces a way to use a sub- 570 field of ID-Reuse without having to set RC=1. In this approach, 571 packet headers using the new protocol will be indistinguishable from 572 an IPv4 header not using the new protocol. Therefore it will be 573 possible to guarantee that middleboxes will not treat packets using 574 the new protocol any differently from other IPv4 packets. 576 Many pre-existing IPv4 hosts cycle through all the values in the IP 577 ID field even when sending atomic packets in which the IP ID field 578 has no function. Therefore, these pre-existing IPv4 hosts will 579 occasionally issue a packet that happens to look as if it is using a 580 codepoint of a new protocol using the IP ID field. Without RC=1, 581 there will be no way to distinguish the two. 583 +------+---------------------+--------------------------+ 584 | | middlebox traversal | new protocol recognition | 585 +------+---------------------+--------------------------+ 586 | RC=0 | Assured | Uncertain | 587 | RC=1 | Uncertain | Assured | 588 +------+---------------------+--------------------------+ 590 Table 2: Tradeoff between deterministic middlebox traversal and 591 deterministic protocol recognition 593 Table 2 shows the tradeoff between using RC=0 or RC=1: 595 RC=0: If an implementation of a new protocol uses RC=0, its packets 596 will traverse middleboxes, but it will suffer a small fraction of 597 false positives when recognising which packets using the new 598 protocol -- occasionally it will mistakenly assume a packet is 599 using the new protocol when it is actually just random noise in 600 the IP ID field from a pre-existing implementation. 602 RC=1: If an implementation of a new protocol uses RC=1, its packets 603 may be black-holed by some middleboxes, but it will be certain 604 which packets use the new protocol and which don't. 606 Nonetheless, a probabilistic protocol that can be deployed may be 607 more useful than a deterministic protocol that cannot. 609 6.1. Process Example with RC=0 611 Figure 6 shows an example of how this approach would work with RC=0. 612 For illustration purposes imagine, as in the previous example in 613 Section 5.2, that an experimental track RFC has been published called 614 Experiment A (ExA) that defines bits 14 to 15 of the ID-Reuse field 615 for atomic packets with RC=1. Now imagine another experimental track 616 RFC has been published called Experiment B (ExB) that defines a use 617 for bits 11 to 13 of the ID-Reuse field, but does not require RC=1. 618 In fact a packet is defined as complying with ExB whether RC=1 or 619 RC=0 (i.e., RC=X, where 'X' means don't care). Figure 7 shows the 620 IANA registry entries for these imaginary sub-fields. 621 0 1 2 3 622 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 623 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 624 | ID-Reuse _____ |X 1 0|0 0 0 0 0 0 0 0 0 0 0 0 0| 625 |0 0 0 0 0 0 0 0 0 0 0| ExB |0 0|0 | | 626 | |1 0 1| | | | 627 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 629 Figure 6: Example of Experimental Reuse of Octets 5 & 6 Without 630 Requiring RC=1 632 The bits shown in each row of Figure 6 define the semantics of the 633 bits shown in the next row down, as follows: 635 o The top row identifies that the packet is atomic. The RC flag is 636 don't care ('X'), so RC does not have to be 1. Implementations 637 can clear RC to 0 to traverse awkward middleboxes, but RC can be 638 set to 1 otherwise. 640 o The middle row shows that an implementation of Experiment B (ExB) 641 has set RC=0. It is also using the ID-Reuse field, so it clears 642 all the bits to zero except those in its own sub-field (bits 643 11-13). It will have registered this experimental use with the 644 IANA as shown in the top example of Figure 7. 646 o The bottom row shows that an implementation of ExB has set its 647 3-bit sub-field to codepoint 101, the meaning of which will have 648 been defined in the RFC specifying the ExB protocol. 650 Note that, the process for using protocol ExB without RC=1 651 (Section 6.2) precludes an implementation from using the ExA protocol 652 in the same packet -- any one packet can only be part of one RC=0 653 protocol at a time. 655 6.2. Process for Using the ID-Reuse Field Without Requiring RC=1 657 This approach SHOULD NOT be used unless the preferred approach 658 (Section 5) is impractical due to middleboxes blocking packets with 659 RC set to 1. 661 To follow this non-preferred approach, the registration with the IANA 662 MUST specify that the sub-field of ID-Reuse is defined for 'RC=X', 663 meaning "don't care", that is RC may be either set or cleared (for an 664 example, see the final bullet of the imaginary registration details 665 in Section 8). The RFC defining the relevant ID-Reuse sub-field MUST 666 also make it clear that the sub-field is defined for either value of 667 the Recycled flag (RC=X) in an atomic IPv4 packet. 669 This approach will not be feasible for all protocols; only those that 670 satisfy the severe constraints laid down below. Otherwise, for 671 protocols that cannot satisfy these prerequisite constraints, the 672 preferred approach in Section 5 wth RC=1 will be the only option. 674 Once a sub-field of the ID-Reuse field has been registered with the 675 IANA, implementations of the protocol can use any of the available 676 codepoints within that sub-field in atomic packets without having to 677 set RC=1, if and only if the following constraints can be satisfied: 679 1. New protocol implementations MUST NOT use RC=0 unless the 680 treatments associated with all the new codepoints are generally 681 benign to packets not taking part in the protocol. 'Benign' 682 means the new protocol SHOULD do no more harm to other packets 683 than previous implementations did. Using the term 'SHOULD' 684 rather than 'MUST' does not completely rule out new protocol 685 proposals that might sometimes introduce slightly more harm, but 686 such proposals will need to give strong justifications 688 2. Implementations MUST clear all the other bits of the ID-Reuse 689 field (except those in the new protocol's sub-field) to zero. 690 Note that this is different to the approach with RC=1, where more 691 than one sub-field at once can be non-zero 693 3. In addition the constraints in Section 5.1 must also be 694 satisfied. 696 Constraint #1 is severe but necessary in order to ensure that a new 697 protocol (e.g. ExB) does not harm atomic packets from pre-existing 698 IPv4 implementations. For example, a receiving implementation of ExB 699 can assume that most packets with all zeros in bits 0-10 and 14-15 700 were deliberately set by another implementation of ExB. But many 701 pre-existing implementations of IPv4 will be cycling (sequentially or 702 randomly) through all the IPID values as they send out packets. 703 Occasionally they will send out a packet that happens to look like it 704 complies with protocol ExB. For the case of ExB with a 3-bit sub- 705 field, such false positives will occur with probability 1 in 2^13 706 (~0.01%). We term this the misrecognition probability. 708 If the new protocol were designed to do harm (e.g. to deprioritise 709 certain packets against others) that would be fine for those packets 710 intended to take part in the new protocol. But it would not be 711 acceptable to harm even a small proportion of packets misrecognised 712 as using the new protocol. This is why the RC=0 approach can only be 713 allowed for a new protocol that is generally benign. 715 Constraint #2 is necessary in order to ensure the misrecognition 716 probability remains low. If only one sub-field is allowed at one 717 time, all the other bits in the ID-Reuse field will have to be zero. 718 This ensures that a pre-existing IPv4 implementation cycling through 719 all the IP ID values will collide less frequently with values used 720 for each new protocol. 722 As already stated (Section 5), each new protocol MUST define the all- 723 zeros codepoint of its sub-field to mean that the new protocol is 724 'turned off'. 726 This arrangement ensures that packets with an IPv4 ID of zero will 727 never collide with a codepoint used by any ID-Reuse scheme, whether 728 RC=0 or RC=1. All zeros was deliberately chosen as the common 729 'turned off' codepoint because some pre-existing implementations have 730 used zero as the default IP ID for atomic packets. 732 In either case, whether the Recycled flag is set or not, a sub-field 733 of the ID-Reuse field MUST be registered with the IANA, initially for 734 experimental use, by referencing the relevant experimental track RFC. 735 This will ensure that experiments with different sub-fields of the 736 ID-Reuse field can proceed in parallel on the public Internet without 737 colliding with each other. The referenced RFC MUST define a coherent 738 process for returning the bits for other uses if the experimental 739 approach does not progress to the standards track. 741 The same sub-field can be used with the same semantics as the 742 experiment progresses, initially with the Recycled flag cleared to 0 743 and later set to 1. And the same protocol semantics can be used 744 whether the proposal is experimental or standards track. Thus, the 745 whole process is designed to: 747 1. allow initial experiments to use RC=0 to traverse non-compliant 748 middleboxes (Section 6); 750 2. then, once sufficient middleboxes forward RC=1 packets, the 751 experiment can either be continued with RC=1 (Section 5); 753 3. or the experiment can progress cleanly to the standards track, 754 while still using the same sub-field but with RC=1; 756 4. or the experiment can be terminated without having wasted any 757 header bits. 759 (Step 1 is only feasible if the extra constraints in Section 6.2 can 760 be satisfied. If not, Step 2 will be the only feasible first step.) 762 For the avoidance of doubt, any use of ID-Reuse, whether experimental 763 or not, is also subject to the general constraints already enumerated 764 in Section 5.1. 766 7. Updates to Existing RFCs 768 Great care has been taken to ensure all the updates defined in this 769 specifications are incrementally deployable. 771 The definition of the RC flag in Section 3 updates the status of this 772 flag that was "reserved, must be zero" in [RFC0791]. The 773 redefinition of the IP Identification field as the ID-Reuse field 774 when an IPv4 packet is atomic also updates RFC791. 776 Updates to existing RFC791 implementations are only REQUIRED if they 777 discard IPv4 packets with RC=1, or change RC from 1 to 0, both of 778 which are misinterpretations of RFC791 anyway. Otherwise, there will 779 be no need to update an RFC791-compliant IPv4 stack until new use(s) 780 for the ID-Reuse field are also specified. 782 The recommendation in Section 4.2 to copy the ID-Reuse field when 783 encapsulating an atomic IPv4 packet with another atomic IPv4 header 784 updates IPv4-in-IPv4 encapsulation specifications [RFC2003] 785 [RFC4301]. These updates to tunnels are likely to be recommended 786 rather than essential for interworking, so they can be implemented as 787 part of routine code maintenance. 789 The ability to redefine the IPv4 ID field of an atomic packet updates 790 [RFC6864], specifically the following two statements no longer apply: 791 "the field's value is defined only when a datagram is actually 792 fragmented" and "IPv4 ID field MUST NOT be used for purposes other 793 than fragmentation and reassembly." Nonetheless, octets 5 & 6 of an 794 atomic packet still MUST NOT be interpreted with the semantics of the 795 Identification field. 797 [RFC2780] provides the IANA with guidelines on allocating values in 798 IP and related headers. The process defined in Section 5 and 799 Section 6 update RFC2780, given ID-Reuse is effectively a new field 800 in the IPv4 header. 802 [RFC4727] defines the processes for experimental use of values in 803 fields in the IP header that are managed by the IANA. The processes 804 defined in Section 5 and Section 6 update RFC4727 to include the new 805 alternative use of the IPv4 ID field as an ID-Reuse field. 807 8. IANA Considerations 809 The IANA is requested to establish a new registry to record 810 allocation of sub-divisions of the ID-Reuse field and to avoid 811 duplicate allocations. The ID-Reuse field is an alternative use of 812 the Identification field of the IPv4 header in atomic packets 813 (Section 3). All 16 bits are available for assignment, either as 814 sub-fields of bits or as sets of codepoints within a sub-field of 815 bits. Each sub-division of the ID-Reuse field MUST be allocated 816 through an IETF Consensus action. The registry MUST then record: 818 Protocol name: the name for the protocol, as used in the RFC 819 defining it 821 RFC: the RFC that defines the semantics of the codepoints used by 822 the protocol 824 Leftmost bit: the leftmost bit allocated, counting from bit 0 at the 825 most significant bit (which is bit 32 of the IPv4 header, counting 826 from 0) 828 No. of bits allocated: the width in bits of the allocated sub-field 830 Codepoint range (optional): The range of codepoints within the 831 assigned sub-field of bits that the protocol uses 833 Sub-field defined if: the precondition for the sub-field to be 834 defined (Section 5). Valid entries MUST include the condition 835 that the packet is atomic and MUST specifiy valid values of the 836 Recycled (RC) flag, either 'RC=1' or 'RC=X', where 'X' means don't 837 care (Section 6). 839 Two example registrations are shown in Figure 7. 841 Protocol name: ExB 842 RFC: BBBB 843 Most significant bit: 11 844 No. of bits allocated: 3 845 Codepoint range: all 846 Sub-field defined if: Atomic packet and RC=X 848 Protocol name: ExA 849 RFC: AAAA 850 Most significant bit: 14 851 No. of bits allocated: 2 852 Codepoint range: all 853 Sub-field defined if: Atomic packet and RC=1 855 Figure 7: Example IANA Registry of Sub-fields of the ID-Reuse Field 857 9. Security Considerations 859 Integrity Checking: This specification make the semantics of octets 860 5 & 6 of the IPv4 header (IP ID or ID-Reuse) depend on the setting 861 of octets 7 & 8 (all the Control Flags and the Fragment Offset 862 field). The IP Authentication Header (AH) [RFC4302] covers octets 863 5 & 6 but not octets 7 & 8. Therefore AH can assure the integrity 864 of the bits in the ID-Reuse field, but it cannot verify whether or 865 not the sender intended those bits to have the semantics of an ID- 866 Reuse field. 868 Any security-sensitive application of the ID-Reuse field will 869 therefore need to provide its own integrity checking of the status 870 of the Control Flags and Fragment Offset. Such a facility would 871 need to take into account that the present specification allows an 872 intermediate node to set the Recycled flag, but not to clear it 873 (Section 4.1). 875 Covert channels: It has always been possible to use bit 48 of the 876 IPv4 header for a 1 bit per packet covert channel, for instance 877 between a network protected by IPsec and an unprotected network. 878 Bit 48 could be covertly toggled to pass messages because it had 879 no function (so no-one would notice any affect on the main 880 communication channel) and it was not covered by IPsec 881 authentication. On the other hand, once alerted to the 882 vulnerability, it has always been easy for an IPsec gateway to 883 spot bit 48 being used as a covert channel, given bit 48 was meant 884 to always be zero. 886 Now that bit 48 has been given a function, it will often no longer 887 be possible for an attacker to toggle it without affecting the 888 main data communication. However, whenever the main communication 889 does not depend on bit 48, it will be easier to for an attacker to 890 toggle it covertly given it will no longer stand out as anomalous 891 behaviour. 893 10. Conclusions 895 This specification builds on recent moves to make the approach to 896 fragmentation in IPv4 more closely match that of IPv6. Already the 897 fields that support fragmentation in the IPv4 header are usually 898 redundant, but unfortunately they are non-optional. 900 This specification makes it possible to reuse the 16 bits of the IPv4 901 ID field when they are not needed for reassembly. The last unused 902 bit in the IPv4 header is used in order to unambiguously flag that 903 the IP ID field has new semantics. The bit is called the Recycled 904 flag, because it allows the IP ID field to be recycled for new 905 purposes when it would otherwise be redundant. Whenever the IP ID 906 field has new semantics, it is termed the ID-Reuse field. 908 The process for redefining the semantics of sub-fields of this ID- 909 Reuse field has been laid down, both for experimental and standards 910 actions. Great care has been taken throughout to ease incremental 911 deployment. The same sub-field can be used with the same semantics 912 as an experiment evolves into a standards action. Initially it is 913 even possible for certain experiments to leave the Recycled flag 914 cleared to zero, in order to traverse any awkward middleboxes that 915 incorrectly discard or normalise packets if the Recycled flag is set. 917 11. Acknowledgements 919 Rob Hancock originally pointed out that code to handle new protocols 920 can tell the machine where to look for the relevant header. Dan Wing 921 pointed out that codepoints, not just whole bits, could be assigned 922 for protocols that are mutually exclusive. 924 Bob Briscoe is partly funded by Trilogy, a research project (ICT- 925 216372) supported by the European Community under its Seventh 926 Framework Programme. 928 Comments Solicited (to be removed by the RFC Editor): 930 Comments and questions are encouraged and very welcome. They can be 931 addressed to the IETF Internet Area working group mailing list 932 , and/or to the author(s). 934 12. Outstanding Issues (to be removed when all resolved) 936 1. ... 938 13. References 940 13.1. Normative References 942 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 943 September 1981. 945 [RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 2003, 946 October 1996. 948 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 949 Requirement Levels", BCP 14, RFC 2119, March 1997. 951 [RFC2780] Bradner, S. and V. Paxson, "IANA Allocation 952 Guidelines For Values In the Internet Protocol and 953 Related Headers", BCP 37, RFC 2780, March 2000. 955 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 956 Internet Protocol", RFC 4301, December 2005. 958 [RFC4302] Kent, S., "IP Authentication Header", RFC 4302, 959 December 2005. 961 [RFC4727] Fenner, B., "Experimental Values In IPv4, IPv6, 962 ICMPv4, ICMPv6, UDP, and TCP Headers", RFC 4727, 963 November 2006. 965 [RFC6864] Touch, J., "Updated Specification of the IPv4 ID 966 Field", RFC 6864, February 2013. 968 13.2. Informative References 970 [Cisco.IPv6Ext] Cisco, "IPv6 Extension Headers Review and 971 Considerations", Cisco Technology White Paper , 972 October 2006, . 976 [Fransson04] Fransson, P. and A. Jonsson, "End-to-end 977 measurements on performance penalties of IPv4 978 options", Luleae University of Technology, Technical 979 Report 2004:03, 2004, . 982 [RFC4821] Mathis, M. and J. Heffner, "Packetization Layer Path 983 MTU Discovery", RFC 4821, March 2007. 985 [RFC4963] Heffner, J., Mathis, M., and B. Chandler, "IPv4 986 Reassembly Errors at High Data Rates", RFC 4963, 987 July 2007. 989 Appendix A. Why More Bits Cannot be Freed (To be Removed by RFC Editor) 991 Given this specification uses the last unassigned bit in the IPv4 992 header, it is worth checking whether it can be used to flag a new use 993 for more than the 16 bits in the IP ID field of atomic packets. 995 IHL: Ideally, the Internet header length field (4 bits) could be 996 made redundant if the length of those IPv4 headers with bit 48 set 997 were redefined to be fixed at 20 octets. Then a similar approach 998 to IPv6 could be taken with the Protocol field redefined as a Next 999 Header field and each extension header specifying its own length. 1001 Unfortunately, although IPv4 options are rarely used and generally 1002 ignored, this idea would not be incrementally deployable. There 1003 are probably billions of pre-existing implementations of the IPv4 1004 stack that will use the IHL field to find the transport protocol 1005 header, without ever looking at bit 48. If the IHL field were 1006 given any other semantics conditional on bit 48 being set, all 1007 these pre-existing stacks would break. 1009 Header Checksum: Ideally, the Header Checksum (16 bits) could be 1010 made redundant in those IPv4 headers with bit 48 set. Then a 1011 similar approach to IPv6 could be taken where the integrity of the 1012 IP header relies on the end-to-end checksum of the transport 1013 protocol, which includes the main fields in the IP header. 1015 Unfortunately, again, this idea would not be incrementally 1016 deployable. Pre-existing implementations of the IPv4 stack might 1017 verify the header checksum without ever looking at bit 48. And 1018 anyway IPv4 stacks on probably every pre-existing router 1019 implementation would update the checksum field without knowing to 1020 check whether bit 48 was set. Therefore if the field were used 1021 for any other purpose than a checksum, it would be impossible to 1022 predict how its value might be changed by a combination of pre- 1023 existing and new stacks. 1025 It is clear that reusing fields other than the IPv4 ID would be 1026 fraught with incremental deployment problems. The reason the IPv4 ID 1027 field can be reused, is that an atomic packet already does not need 1028 an Identification field, whether bit 48 is set or not. Setting bit 1029 48 merely allows new implementations that understand ID-Reuse 1030 semantics to be certain the value in the ID-Reuse field was not 1031 written by an implementation that intended it to have Identification 1032 semantics. 1034 Appendix B. Experimental or Standards Track? (To Be Removed Before 1035 Publication) 1037 This document defines a protocol (using the Recycled flag) to enable 1038 other protocols (using the ID-Reuse field). The Recycled flag 1039 protocol is currently written as if it is on the IETF standards 1040 track. Nonetheless it might be feasible to write it for the 1041 experimental track. This appendix discusses the pros and cons of 1042 each. 1044 The Recycled flag uses up the last unused bit in the IPv4 header. 1045 The present specification defines a use for this last bit in the 1046 expectation that the Internet community will find ingeneous new 1047 use(s) for sub-fields of the ID-Reuse field, because then the 1048 Recycled flag will be needed to unambiguously indicate the new 1049 semantics. However, there is a risk that the last IPv4 header bit 1050 could be wasted, if no new uses for the IP ID field can be found 1051 within the constraints of its previous use for fragment reassembly, 1052 or if new experimental uses are proposed but none successfully 1053 proceed through to standards actions. 1055 The risk of wasting the last bit would be mitigated if the definition 1056 of the Recycled flag itself was initially on the experimental track. 1057 Then, if some experimental use(s) of the ID-Reuse field did see 1058 widespread adoption, the RC flag protocol could progress to the 1059 standards track. On the other hand, if no ID-Reuse experiments 1060 happened, the RC flag could possibly be reclaimed for another use in 1061 the future. This would require all experiments with the RC flag to 1062 be confined in time, so that stray implementations of old experiments 1063 would not conflict with future uses of the flag. 1065 Eventually, each specification for each sub-field of ID-Reuse might 1066 either progress on the experimental track or standards track. 1067 However, an enabler for standards track specifications cannot itself 1068 only be experimental. Therefore the RC flag protocol would have to 1069 be on the standards track, to enable standards track protocols as 1070 well as experimental. Figure 8 illustrates this need for the RC flag 1071 protocol to have sufficient rank for any protocols it enables. 1073 +---------+---------------------------+ 1074 | RC flag | ID-Reuse sub-field track | 1075 | track +-------------+-------------+ 1076 | | Expt | Stds | 1077 +---------+-------------+-------------+ 1078 | Expt | Expt | INVALID | 1079 | Stds | Expt | Stds | 1080 +---------+-------------+-------------+ 1082 The IETF track of the RC flag protocol in the present document (rows) 1083 and of any particular RFC specifying a sub-field of the ID-Reuse 1084 field (columns). The combination determines the status of any 1085 particular sub-field as shown at the intersection of the relevant row 1086 and column. 1088 Figure 8: Validity of Combinations of IETF tracks for the RC flag and 1089 an ID-Reuse Subfield 1091 One purpose of the present draft is to outline how new uses of ID- 1092 Reuse sub-fields can progress seamlessly from experimental track to 1093 standards track. Therefore, this draft is written as if it were on 1094 the standards track. Otherwise the processes for enabling standards 1095 track documents would have had to be written hypothetically, which 1096 would have been highly confusing. Nonetheless, no intent to prejudge 1097 that this document should be or will be on the standards track is 1098 implied. 1100 If it were decided that the present draft should start on the 1101 experimental track, all the text about enabling standards track 1102 protocols would have to be edited out, or perhaps moved to a non- 1103 normative appendix. 1105 Alternatively, the IETF might see some obvious new uses for sub- 1106 fields of the ID-Reuse field that would make it reasonable to fast- 1107 track the RC flag straight onto the standards track. 1109 Appendix C. Changes in This Version (to be removed by RFC Editor) 1111 From briscoe-03 to 04: 1113 * Refreshed to keep alive. 1115 From briscoe-02 to 03: 1117 * Updated references 1119 From briscoe-01 to 02: 1121 * Altered summary of draft-ietf-ipv4-id-update to reflect recent 1122 changes to that draft 1124 * Added FAQ2 explaining why it will still sometimes be necessary 1125 to update IPv4 even though the focus of the new features will 1126 be IPv6 1128 * Updated references 1130 From briscoe-00 to 01: 1132 * Updated to preserve liveness. 1134 * No changes other than updates to refs and minor corrections. 1136 Author's Address 1138 Bob Briscoe 1139 BT 1140 B54/77, Adastral Park 1141 Martlesham Heath 1142 Ipswich IP5 3RE 1143 UK 1145 Phone: +44 1473 645196 1146 EMail: bob.briscoe@bt.com 1147 URI: http://bobbriscoe.net/