idnits 2.17.1 draft-ietf-pcp-server-selection-07.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 -- The document date (December 17, 2014) is 3417 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 (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 PCP Working Group M. Boucadair 3 Internet-Draft France Telecom 4 Updates: 6887 (if approved) R. Penno 5 Intended status: Standards Track D. Wing 6 Expires: June 20, 2015 P. Patil 7 T. Reddy 8 Cisco 9 December 17, 2014 11 PCP Server Selection 12 draft-ietf-pcp-server-selection-07 14 Abstract 16 The document specifies the behavior to be followed by a PCP client to 17 contact its PCP server(s) when one or several PCP server IP addresses 18 are configured. 20 This document updates RFC6887. 22 Requirements Language 24 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 25 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 26 document are to be interpreted as described in RFC 2119 [RFC2119]. 28 Status of This Memo 30 This Internet-Draft is submitted in full conformance with the 31 provisions of BCP 78 and BCP 79. 33 Internet-Drafts are working documents of the Internet Engineering 34 Task Force (IETF). Note that other groups may also distribute 35 working documents as Internet-Drafts. The list of current Internet- 36 Drafts is at http://datatracker.ietf.org/drafts/current/. 38 Internet-Drafts are draft documents valid for a maximum of six months 39 and may be updated, replaced, or obsoleted by other documents at any 40 time. It is inappropriate to use Internet-Drafts as reference 41 material or to cite them other than as "work in progress." 43 This Internet-Draft will expire on June 20, 2015. 45 Copyright Notice 47 Copyright (c) 2014 IETF Trust and the persons identified as the 48 document authors. All rights reserved. 50 This document is subject to BCP 78 and the IETF Trust's Legal 51 Provisions Relating to IETF Documents 52 (http://trustee.ietf.org/license-info) in effect on the date of 53 publication of this document. Please review these documents 54 carefully, as they describe your rights and restrictions with respect 55 to this document. Code Components extracted from this document must 56 include Simplified BSD License text as described in Section 4.e of 57 the Trust Legal Provisions and are provided without warranty as 58 described in the Simplified BSD License. 60 Table of Contents 62 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 63 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 3 64 3. IP Address Selection: PCP Server with Multiple IP Addresses . 3 65 4. IP Address Selection: Multiple PCP Servers . . . . . . . . . 4 66 5. Example: Multiple PCP Servers on a Single Interface . . . . . 5 67 6. Security Considerations . . . . . . . . . . . . . . . . . . . 6 68 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 7 69 8. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 7 70 9. References . . . . . . . . . . . . . . . . . . . . . . . . . 7 71 9.1. Normative References . . . . . . . . . . . . . . . . . . 7 72 9.2. Informative References . . . . . . . . . . . . . . . . . 7 73 Appendix A. Multi-homing . . . . . . . . . . . . . . . . . . . . 8 74 A.1. IPv6 Multi-homing . . . . . . . . . . . . . . . . . . . . 8 75 A.2. IPv4 Multi-homing . . . . . . . . . . . . . . . . . . . . 9 77 1. Introduction 79 A host may have multiple network interfaces (e.g., 3G, IEEE 802.11, 80 etc.); each configured with different PCP servers. Each PCP server 81 learned must be associated with the interface on which it was 82 learned. Generic multi-interface considerations are documented in 83 Section 8.4 of [RFC6887]. Multiple PCP server IP addresses may be 84 configured on a PCP client in some deployment contexts such as multi- 85 homing (see Appendix A). A PCP server may also have multiple IP 86 addresses associated with it. It is out of scope of this document to 87 enumerate all deployment scenarios that require multiple PCP server 88 IP addresses to be configured. 90 If a PCP client discovers multiple PCP server IP addresses, it needs 91 to determine which actions it needs to undertake (e.g., whether PCP 92 entries are to be installed in all or a subset of discovered IP 93 addresses, whether some PCP entries are to be removed, etc.). This 94 document makes the following assumptions: 96 o There is no requirement that multiple PCP servers configured on 97 the same interface have the same capabilities. 99 o PCP requests to different PCP servers are independent, the result 100 of a PCP request to one PCP server does not influence another. 102 o The configuration mechanism must distinguish IP addresses that 103 belong to the same PCP server. 105 This document specifies the behavior to be followed by a PCP client 106 [RFC6887] to contact its PCP server(s) [RFC6887] when it is 107 configured with one or several PCP server IP addresses (e.g., using 108 DHCP [RFC7291]). 110 2. Terminology 112 This document makes use of the following terms: 114 o PCP client: denotes a PCP software instance responsible for 115 issuing PCP requests to a PCP server. Refer to [RFC6887]. 116 o PCP server: denotes a software instance that receives and 117 processes PCP requests from a PCP client. A PCP server can be co- 118 located with or be separated from the function it controls (e.g., 119 Network Address Translation (NAT) or firewall). Refer to 120 [RFC6887]. 122 3. IP Address Selection: PCP Server with Multiple IP Addresses 124 This section describes the behavior a PCP client follows to contact 125 its PCP server when the PCP client has multiple IP addresses for a 126 single PCP server. 128 1. A PCP client should construct a set of candidate source addresses 129 (Section 4 of [RFC6724]), based on application input and PCP 130 [RFC6887] constraints. For example, when sending a PEER or a MAP 131 with FILTER request for an existing TCP connection, the only 132 candidate source address is the source address used for the 133 existing TCP connection. But when sending a MAP request for a 134 service that will accept incoming connections, the candidate 135 source addresses may be all of the node's IP addresses, or some 136 subset of IP addresses on which the service is configured to 137 listen. 139 2. The PCP client then sorts the PCP server IP addresses as per 140 Section 6 of [RFC6724] using the candidate source addresses 141 selected in the previous step as input to the destination address 142 selection algorithm. 144 3. The PCP client initializes its Maximum Retransmission Count (MRC) 145 to 4. 147 4. The PCP client sends its PCP messages following the 148 retransmission procedure specified in Section 8.1.1 of [RFC6887]. 149 If no response is received after MRC attempts, the PCP client re- 150 tries the procedure with the next IP address in the sorted list. 151 The PCP client SHOULD ignore any response received from an IP 152 address after exhausting MRC attempts for that particular IP 153 address. If, when sending PCP requests, the PCP client receives 154 a hard ICMP error [RFC1122] it SHOULD immediately try the next IP 155 address from the list of PCP server IP addresses. 157 5. If the PCP client has exhausted all IP addresses configured for a 158 given PCP server, the procedure SHOULD be repeated every fifteen 159 (15) minutes until the PCP request is successfully answered. 161 6. Once the PCP client has successfully received a response from a 162 PCP server's IP address, all subsequent PCP requests to that PCP 163 server are sent on the same IP address until that IP address 164 becomes unresponsive. In case the IP address becomes 165 unresponsive, the PCP client clears the cache of sorted 166 destination addresses and follows the steps described above to 167 contact the PCP server again. 169 For efficiency, the PCP client SHOULD use the same Mapping Nonce for 170 requests sent to all IP addresses belonging to the same PCP server. 172 4. IP Address Selection: Multiple PCP Servers 174 This section describes the behavior a PCP client follows to contact 175 multiple PCP servers, with each PCP server reachable on a list of IP 176 addresses. There is no requirement that these multiple PCP servers 177 have the same capabilities. 179 If several PCP servers are configured, each with multiple IP 180 addresses, the PCP client contacts all PCP servers using the 181 procedure described in Section 3. 183 As specified in Section 11.2 and Section 12.2 of [RFC6887], the PCP 184 client must use a different Mapping Nonce for each PCP server it 185 communicates with. 187 If the PCP client is configured, using some means, with the 188 capabilities of each PCP server, a PCP client may choose to contact 189 all PCP servers simultaneously or iterate through them with a delay. 191 This procedure may result in a PCP client instantiating multiple 192 mappings maintained by distinct PCP servers. The decision to use all 193 these mappings or delete some of them depends on the purpose of the 194 PCP request. For example, if the PCP servers are configuring 195 firewall (not NAT) functionality then the client would by default 196 (i.e., unless it knows that they all replicate state among them) need 197 to use all the PCP servers. 199 5. Example: Multiple PCP Servers on a Single Interface 201 Figure 1 depicts an example that is used to illustrate the server 202 selection procedure specified in Section 3 and Section 4. In this 203 example, PCP servers (A and B) are co-located with edge routers 204 (rtr1, rtr2) with each PCP server controlling its own device. 206 ISP Network 207 | | 208 ......................................................... 209 | | Subscriber Network 210 +-------+--------+ +----+-----------+ 211 | PCP-Server-A | | PCP-Server-B | 212 | (rtr1) | | (rtr2) | 213 +-------+--------+ +----+-----------+ 214 192.0.2.1 | | 198.51.100.1 215 2001:db8:1111::1 | | 2001:db8:2222::1 216 | | 217 | | 218 -------+--------------+----------- 219 | 220 | 203.0.113.0 221 | 2001:db8:3333::1 222 +---+---+ 223 | Host | 224 +-------+ 226 Edge Routers (rtr1, rtr2) 228 Figure 1 230 The example describes behavior when a single IP address for one PCP 231 server is not responsive. The PCP client is configured with two PCP 232 servers for the same interface, PCP-Server-A and PCP-Server-B each 233 having two IP addresses, an IPv4 address and an IPv6 address. The 234 PCP client wants an IPv4 mapping so it orders the addresses as 235 follows: 237 o PCP-Server-A: 239 * 192.0.2.1 241 * 2001:db8:1111::1 243 o PCP-Server-B: 245 * 198.51.100.1 247 * 2001:db8:2222::1 249 Suppose that: 251 o The path to reach 192.0.2.1 is broken 253 o The path to reach 2001:db8:1111::1 is working 255 o The path to reach 198.51.100.1 is working 257 o The path to reach 2001:db8:2222::1 is working 259 It sends two PCP requests at the same time, the first to 192.0.2.1 260 (corresponding to PCP-Server-A) and the second to 198.51.100.1 261 (corresponding to PCP-Server-B). The path to 198.51.100.1 is working 262 so a PCP response is received. Because the path to 192.0.2.1 is 263 broken, no PCP response is received. The PCP client retries 4 times 264 to elicit a response from 192.0.2.1 and finally gives up on that 265 address and sends a PCP message to 2001::db8:1111:1. That path is 266 working, and a response is received. Thereafter, the PCP client 267 should continue using that responsive IP address for PCP-Server-A 268 (2001:db8:1111::1). In this particular case, it will have to use 269 THIRD_PARTY option for IPv4 mappings. 271 6. Security Considerations 273 PCP related security considerations are discussed in [RFC6887]. 275 This document does not specify how PCP server addresses are 276 provisioned on the PCP client. It is the responsibility of PCP 277 server provisioning document(s) to elaborate on security 278 considerations to discover legitimate PCP servers. 280 7. IANA Considerations 282 This document does not request any action from IANA. 284 8. Acknowledgements 286 Many thanks to Dave Thaler, Simon Perreault, Hassnaa Moustafa, and 287 Ted Lemon for their reviews and comments. 289 9. References 291 9.1. Normative References 293 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 294 Requirement Levels", BCP 14, RFC 2119, March 1997. 296 [RFC6724] Thaler, D., Draves, R., Matsumoto, A., and T. Chown, 297 "Default Address Selection for Internet Protocol Version 6 298 (IPv6)", RFC 6724, September 2012. 300 [RFC6887] Wing, D., Cheshire, S., Boucadair, M., Penno, R., and P. 301 Selkirk, "Port Control Protocol (PCP)", RFC 6887, April 302 2013. 304 9.2. Informative References 306 [RFC1122] Braden, R., "Requirements for Internet Hosts - 307 Communication Layers", STD 3, RFC 1122, October 1989. 309 [RFC4116] Abley, J., Lindqvist, K., Davies, E., Black, B., and V. 310 Gill, "IPv4 Multihoming Practices and Limitations", RFC 311 4116, July 2005. 313 [RFC7291] Boucadair, M., Penno, R., and D. Wing, "DHCP Options for 314 the Port Control Protocol (PCP)", RFC 7291, July 2014. 316 Appendix A. Multi-homing 318 The main problem of a PCP multi-homing situation can be succinctly 319 described as 'one PCP client, multiple PCP servers'. As described in 320 Section 3, if a PCP client discovers multiple PCP servers, it should 321 send requests to all of them with assumptions described in Section 1. 323 The following sub-sections describe multi-homing examples to 324 illustrate the PCP client behavior. 326 A.1. IPv6 Multi-homing 328 In this example of an IPv6 multi-homed network, two or more routers 329 co-located with firewalls are present on a single link shared with 330 the host(s). Each router is in turn connected to a different service 331 provider network and the host in this environment would be offered 332 multiple prefixes and advertised multiple DNS servers. Consider a 333 scenario in which firewalls within an IPv6 multi-homing environment 334 also implement a PCP server. The PCP client learns the available PCP 335 servers using DHCP [RFC7291] or any other provisioning mechanism. In 336 reference to Figure 2, a typical model is to embed DHCP servers in 337 rtr1 and rtr2. A host located behind rtr1 and rtr2 can contact these 338 two DHCP servers and retrieve from each server the IP address(es) of 339 the corresponding PCP server. 341 The PCP client will send PCP requests in parallel to each of the PCP 342 servers. 344 ================== 345 | Internet | 346 ================== 347 | | 348 | | 349 +----+-+ +-+----+ 350 | ISP1 | | ISP2 | 351 +----+-+ +-+----+ ISP Network 352 | | 353 ......................................................... 354 | | 355 | | Subscriber Network 356 +-------+---+ +----+------+ 357 | rtr1 with | | rtr2 with | 358 | FW1 | | FW2 | 359 +-------+---+ +----+------+ 360 | | 361 | | 362 -------+----------+------ 363 | 364 +---+---+ 365 | Host | 366 +-------+ 368 Figure 2: IPv6 Multihoming 370 A.2. IPv4 Multi-homing 372 In this example an IPv4 multi-homed network described in 'NAT- or 373 RFC2260-based multi-homing' (Section 3.3 of [RFC4116]), the gateway 374 router is connected to different service provider networks. This 375 method uses Provider-Aggregatable (PA) addresses assigned by each 376 transit provider to which the site is connected. The site uses NAT 377 to translate the various provider addresses into a single set of 378 private-use addresses within the site. In such a case, two PCP 379 servers might have to be present to configure NAT to each of the 380 transit providers. The PCP client learns the available PCP servers 381 using DHCP [RFC7291] or any other provisioning mechanism. In 382 reference to Figure 3, a typical model is to embed the DHCP server 383 and the PCP servers in rtr1. A host located behind rtr1 can contact 384 the DHCP server to obtain IP addresses of the PCP servers. The PCP 385 client will send PCP requests in parallel to each of the PCP servers. 387 ===================== 388 | Internet | 389 ===================== 390 | | 391 | | 392 +----+--------+ +-+------------+ 393 | ISP1 | | ISP2 | 394 | | | | 395 +----+--------+ +-+------------+ ISP Network 396 | | 397 | | 398 .............................................................. 399 | | 400 | Port1 | Port2 Subscriber Network 401 | | 402 +----+-------------------+ 403 |rtr1: NAT & PCP servers | 404 | GW Router | 405 +----+-------------------+ 406 | 407 | 408 | 409 -----+-------------- 410 | 411 +-+-----+ 412 | Host | (private address space) 413 +-------+ 415 Figure 3: IPv4 Multihoming 417 Authors' Addresses 419 Mohamed Boucadair 420 France Telecom 421 Rennes 35000 422 France 424 EMail: mohamed.boucadair@orange.com 426 Reinaldo Penno 427 Cisco 428 USA 430 EMail: repenno@cisco.com 431 Dan Wing 432 Cisco Systems, Inc. 433 170 West Tasman Drive 434 San Jose, California 95134 435 USA 437 EMail: dwing@cisco.com 439 Prashanth Patil 440 Cisco Systems, Inc. 441 Bangalore 442 India 444 EMail: praspati@cisco.com 446 Tirumaleswar Reddy 447 Cisco Systems, Inc. 448 Cessna Business Park, Varthur Hobli 449 Sarjapur Marathalli Outer Ring Road 450 Bangalore, Karnataka 560103 451 India 453 EMail: tireddy@cisco.com