idnits 2.17.1 draft-ietf-6man-impatient-nud-03.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 : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year (Using the creation date from RFC4861, updated by this document, for RFC5378 checks: 2004-07-16) -- 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 (Oct 2012) is 4183 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) No issues found here. Summary: 0 errors (**), 0 flaws (~~), 1 warning (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 6MAN WG E. Nordmark 3 Internet-Draft Cisco Systems, Inc. 4 Updates: 4861 (if approved) I. Gashinsky 5 Intended status: Standards Track Yahoo! 6 Expires: April 4, 2013 Oct 2012 8 Neighbor Unreachability Detection is too impatient 9 draft-ietf-6man-impatient-nud-03.txt 11 Abstract 13 IPv6 Neighbor Discovery includes Neighbor Unreachability Detection. 14 That function is very useful when a host has an alternative, for 15 instance multiple default routers, since it allows the host to switch 16 to the alternative in short time. This time is 3 seconds after the 17 node starts probing by default. However, if there are no 18 alternatives, this is far too impatient. This document specifies 19 relaxed rules for Neighbor Discovery retransmissions that allows an 20 implementation to choose different timeout behavior based on whether 21 or not there are alternatives. This document updates RFC 4861. 23 Status of this Memo 25 This Internet-Draft is submitted in full conformance with the 26 provisions of BCP 78 and BCP 79. 28 Internet-Drafts are working documents of the Internet Engineering 29 Task Force (IETF). Note that other groups may also distribute 30 working documents as Internet-Drafts. The list of current Internet- 31 Drafts is at http://datatracker.ietf.org/drafts/current/. 33 Internet-Drafts are draft documents valid for a maximum of six months 34 and may be updated, replaced, or obsoleted by other documents at any 35 time. It is inappropriate to use Internet-Drafts as reference 36 material or to cite them other than as "work in progress." 38 This Internet-Draft will expire on April 4, 2013. 40 Copyright Notice 42 Copyright (c) 2012 IETF Trust and the persons identified as the 43 document authors. All rights reserved. 45 This document is subject to BCP 78 and the IETF Trust's Legal 46 Provisions Relating to IETF Documents 47 (http://trustee.ietf.org/license-info) in effect on the date of 48 publication of this document. Please review these documents 49 carefully, as they describe your rights and restrictions with respect 50 to this document. Code Components extracted from this document must 51 include Simplified BSD License text as described in Section 4.e of 52 the Trust Legal Provisions and are provided without warranty as 53 described in the Simplified BSD License. 55 Table of Contents 57 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 58 2. Definition Of Terms . . . . . . . . . . . . . . . . . . . . . . 4 59 3. Protocol Updates . . . . . . . . . . . . . . . . . . . . . . . 4 60 4. Example Algorithm . . . . . . . . . . . . . . . . . . . . . . . 6 61 5. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 7 62 6. Security Considerations . . . . . . . . . . . . . . . . . . . . 7 63 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . . 7 64 8. References . . . . . . . . . . . . . . . . . . . . . . . . . . 7 65 8.1. Normative References . . . . . . . . . . . . . . . . . . . 7 66 8.2. Informative References . . . . . . . . . . . . . . . . . . 8 67 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 8 69 1. Introduction 71 IPv6 Neighbor Discovery [RFC4861] includes Neighbor Unreachability 72 Detection (NUD), which detects when a neighbor is no longer 73 reachable. The timeouts specified are very short (by default three 74 transmissions spaced one second apart). That can be appropriate when 75 there are alternative paths over which the packets can be sent. For 76 example, if a host has multiple default routers in its Default Router 77 List, or if the host has a Neighbor Cache Entry (NCE) created by a 78 Redirect message. The effect of NUD reporting a failure in those 79 cases is that the host will try the alternative; the next router in 80 the Default Router List, or discard the NCE which will also send 81 using a different router. 83 For that reason the timeouts in [RFC4861] were chosen to be short; 84 this ensures that if a default router fails the host can use the next 85 router in less than 45 seconds (taking into account a default 86 ReachableTime of 30 seconds and the time spent in the DELAY state). 88 However, when there is no alternative there are several benefits in 89 making NUD try probing for a longer time. One of those benefits is 90 to be more robust against transient failures, such as spanning tree 91 reconvergence and other layer 2 issues that can take many seconds to 92 resolve. Marking the NCE as unreachable in that case causes 93 additional multicast on the network. Assuming there are IP packets 94 to send, the lack of an NCE will result in multicast Neighbor 95 Solicitations every second instead of the unicast Neighbor 96 Solicitations that NUD sends. 98 As a result IPv6 Neighbor Discovery is operationally more brittle 99 than IPv4 ARP. For IPv4 there is no mandatory time limit on the 100 retransmission behavior for ARP [RFC0826] which allows implementors 101 to pick more robust schemes. 103 The following constant values in [RFC4861] seem to have been made 104 part of IPv6 conformance testing: MAX_MULTICAST_SOLICIT, 105 MAX_UNICAST_SOLICIT, and RETRANS_TIMER. While such strict 106 conformance testing seems consistent with [RFC4861], it means that we 107 need to update the standard if we want to allow IPv6 Neighbor 108 Discovery to be as robust as ARP. 110 This document updates RFC 4861 to relax the retransmission rules. 112 Additional motivations for making IPv6 Neighbor Discovery more robust 113 in the face of degenerate conditions are covered in [RFC6583]. 115 2. Definition Of Terms 117 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 118 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 119 document are to be interpreted as described in [RFC2119]. 121 3. Protocol Updates 123 Giving up after three packets spaced one second apart is only 124 REQUIRED when there is an alternative, such as an additional default 125 router or a redirect. 127 If implementations transmit more than MAX_*CAST_SOLICIT packets it 128 SHOULD use (binary) exponential backoff of the retransmit timer. 129 This is to avoid any significant load due to a steady background 130 level of retransmissions from implementations that try for a long 131 time. 133 Even if there is no alternative, the protocol needs to be able to 134 handle the case when the link-layer address of the destination has 135 changed by switching to multicast Neighbor Solicitations at some 136 point in time. 138 In order to capture all the cases above this document introduces a 139 new UNREACHABLE state in the conceptual model described in [RFC4861]. 140 A NCE in the UNREACHABLE state retains the link-layer address, and 141 IPv6 packets continue to be sent to that link-layer address. But in 142 the UNREACHABLE state the NUD Neighbor Solicitations are multicast, 143 using a timeout that follows a (binary) exponential backoff. 145 In the places where RFC4861 says to to discard/delete the NCE after N 146 probes (Section 7.3, 7.3.3 and Appendix C) we will instead transition 147 to the UNREACHABLE state. 149 If the Neighbor Cache Entry was created by a redirect, a node MAY 150 delete the NCE instead of changing its state to UNREACHABLE. In any 151 case, the node SHOULD NOT use an NCE created by a Redirect to send 152 packets if that NCE is in unreachable state. Packets should be sent 153 following the next-hop selection algorithm in section 5.2 in 154 [RFC4861] which disregards NCEs that are not reachable. 156 The default router selection in section 6.3.6 says to prefer default 157 routers that are "known to be reachable". For the purposes of that 158 section, if the NCE for the router is in UNREACHABLE state, it is not 159 known to be reachable. Thus the particular text in section 6.3.6 160 which says "in any state other than INCOMPLETE" needs to be extended 161 to say "in any state other than INCOMPLETE or UNREACHABLE". 163 Apart from the use of multicast NS instead of unicast NS, and the 164 (binary) exponential backoff of the timer, the UNREACHABLE state 165 works the same as the current PROBE state. 167 A node MAY garbage collect a Neighbor Cache Entry at any time as 168 specified in RFC 4861. This does not change with the introduction of 169 the UNREACHABLE state in the conceptual model. 171 The UNREACHABLE state is conceptual and not a required part of this 172 specification. Just as for [RFC4861] a node merely needs to satisfy 173 the externally observable behavior of this specification. 175 There is a non-obvious extension to the state machine description in 176 Appendix C in RFC 4861 in the case for "NA, Solicited=1, Override=0. 177 Different link-layer address than cached". There we need to add 178 "UNREACHABLE" to the current list of "STALE, PROBE, Or DELAY". That 179 is, the NCE would be unchanged. Note that there is no corresponding 180 change necessary to the text in section 7.2.5 since it is phrased 181 using "Otherwise" instead of explicitly listing the three states. 183 The other state transitions described in Appendix C handle the 184 introduction of the UNREACHABLE state without any change, since they 185 are described using "not INCOMPLETE". 187 There is also the more obvious change already described above. RFC 188 4861 has this: 190 PROBE Retransmit timeout, Discard entry - 191 N or more 192 retransmissions. 194 That needs to be replaced by: 196 PROBE Retransmit timeout, Double timeout UNREACHABLE 197 N or more Send multicast NS 198 retransmissions. 200 UNREACHABLE Retransmit timeout Double timeout UNREACHABLE 201 Send multicast NS 203 The binary exponential backoff SHOULD be clamped at some reasonable 204 maximum retransmit timeout, such as 60 seconds (MAX_RETRANS_TIMER 205 below). If there is no IPv6 packet sent using the UNREACHABLE NCE, 206 then it makes sense to stop the retransmits of the multicast NS until 207 either the NCE is garbage collected or there are IPv6 packets sent 208 using the NCE. The multicast NS and associated binary exponential 209 backoff can be applied on the condition of the continued use of the 210 NCE to send IPv6 packets to the recorded link-layer address. 212 A node MAY unicast the first few Neighbor Solicitation messages while 213 in UNREACHABLE state, but it MUST switch to multicast Neighbor 214 Solicitations. Otherwise it would not detect a link-layer address 215 change for the target. 217 4. Example Algorithm 219 This section is NOT normative, but specifies a simple implementation 220 which conforms with this document. The implementation is described 221 using operator configurable values that allows it to be configured in 222 a way to be compatible with the retransmission behavior in [RFC4861]. 223 The operator can configure the values for MAX_*CAST_SOLICIT, 224 RETRANS_TIMER, and the new BACKOFF_MULTIPLE and MARK_UNREACHABLE. 225 This allows the implementation to be as simple as: 227 next_retrans = ($BACKOFF_MULTIPLE^$solicit_attempt_num)*$RetransTimer 228 + jittered value. 230 After MARK_UNREACHABLE retransmissions the implementation would mark 231 the NCE UNREACHABLE and switch to multicast NUD probes. 233 The recommended behavior is to have 5 attempts, with timing spacing 234 of 0 (initial request), 1 second later, 3 seconds later, then 9, then 235 27, then 60, then 60 and switch to UNREACHABLE after the three first 236 transmissions. Thus relative to the time of the first transmissions 237 the retransmissions would occur at 1 second, 4 seconds, 13 seconds, 238 40 seconds, then 100 seconds, and finally 160 seconds. At 4 seconds 239 from the first transmission the NCE would be marked UNREACHABLE. 240 That recommended behavior corresponds to: 242 MAX_UNICAST_SOLICIT=5 244 RETRANS_TIMER=1 (default) 246 MAX_RETRANS_TIMER=60 248 BACKOFF_MULTIPLE=3 250 MARK_UNREACHABLE=3 252 After 3 retransmissions the implementation would mark the NCE 253 UNREACHABLE and switch to multicast NUD probes. Thus we enter 254 UNREACHABLE, and try any available alternative, after 4 seconds 255 compared to the current 2 seconds. That additional delay is small 256 compared to the default 30 seconds ReachableTime. 258 If BACKOFF_MULTIPLE=1, MARK_UNREACHABLE=3 and MAX_UNICAST_SOLICIT=3, 259 you would get the same behavior as in [RFC4861]. 261 An implementation following this algorithm would, if the request was 262 not answered at first due for example to a transitory condition, 263 retry immediately, and then back off for progressively longer 264 periods. This would allow for a reasonably fast resolution time when 265 the transitory condition clears. 267 Note that RetransTimer and ReachableTime are by default set from the 268 protocol constants RETRANS_TIMER and REACHABLE_TIME, but are 269 overridden by values advertised in Router Advertisements as specified 270 in [RFC4861]. That remains the case even with the protocol updates 271 specified in this document. The key values that the operator would 272 configure are BACKOFF_MULTIPLE, MAX_UNICAST_SOLICIT and 273 MAX_MULTICAST_SOLICIT. 275 It would be useful to have a maximum value for 276 ($BACKOFF_MULTIPLE^$solicit_attempt_num)*$RetransTimer so that the 277 retransmissions are not too far apart. A value 60 seconds is 278 consistent with DHCP. 280 5. Acknowledgements 282 The comments from Thomas Narten, Philip Homburg, Joel Jaeggli, and 283 Hemant Singh have helped improve this draft. 285 6. Security Considerations 287 Relaxing the retransmission behavior for NUD is believed to have no 288 impact on security. In particular, it doesn't impact the application 289 Secure Neighbor Discovery [RFC3971]. 291 7. IANA Considerations 293 This are no IANA considerations for this document. 295 8. References 297 8.1. Normative References 299 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 300 Requirement Levels", BCP 14, RFC 2119, March 1997. 302 [RFC3971] Arkko, J., Kempf, J., Zill, B., and P. Nikander, "SEcure 303 Neighbor Discovery (SEND)", RFC 3971, March 2005. 305 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 306 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 307 September 2007. 309 8.2. Informative References 311 [RFC0826] Plummer, D., "Ethernet Address Resolution Protocol: Or 312 converting network protocol addresses to 48.bit Ethernet 313 address for transmission on Ethernet hardware", STD 37, 314 RFC 826, November 1982. 316 [RFC6583] Gashinsky, I., Jaeggli, J., and W. Kumari, "Operational 317 Neighbor Discovery Problems", RFC 6583, March 2012. 319 Authors' Addresses 321 Erik Nordmark 322 Cisco Systems, Inc. 323 510 McCarthy Blvd. 324 Milpitas, CA, 95035 325 USA 327 Phone: +1 408 527 6625 328 Email: nordmark@cisco.com 330 Igor Gashinsky 331 Yahoo! 332 45 W 18th St 333 New York, NY 334 USA 336 Email: igor@yahoo-inc.com