idnits 2.17.1 draft-ietf-sidr-bgpsec-protocol-18.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 seems to use 'NOT RECOMMENDED' as an RFC 2119 keyword, but does not include the phrase in its RFC 2119 key words list. -- The document date (August 18, 2016) is 2807 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(See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Outdated reference: A later version (-36) exists of draft-ietf-idr-bgp-extended-messages-13 == Outdated reference: A later version (-18) exists of draft-ietf-sidr-bgpsec-algs-15 == Outdated reference: A later version (-21) exists of draft-ietf-sidr-bgpsec-pki-profiles-18 == Outdated reference: A later version (-06) exists of draft-ietf-sidr-as-migration-05 == Outdated reference: A later version (-16) exists of draft-ietf-sidr-bgpsec-ops-10 == Outdated reference: A later version (-09) exists of draft-ietf-sidr-rpki-rtr-rfc6810-bis-07 == Outdated reference: A later version (-16) exists of draft-ietf-sidr-rtr-keying-12 Summary: 0 errors (**), 0 flaws (~~), 9 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group M. Lepinski, Ed. 3 Internet-Draft NCF 4 Intended status: Standards Track K. Sriram, Ed. 5 Expires: February 19, 2017 NIST 6 August 18, 2016 8 BGPsec Protocol Specification 9 draft-ietf-sidr-bgpsec-protocol-18 11 Abstract 13 This document describes BGPsec, an extension to the Border Gateway 14 Protocol (BGP) that provides security for the path of autonomous 15 systems through which a BGP update message passes. BGPsec is 16 implemented via an optional non-transitive BGP path attribute that 17 carries a digital signature produced by each autonomous system that 18 propagates the update message. 20 Status of This Memo 22 This Internet-Draft is submitted in full conformance with the 23 provisions of BCP 78 and BCP 79. 25 Internet-Drafts are working documents of the Internet Engineering 26 Task Force (IETF). Note that other groups may also distribute 27 working documents as Internet-Drafts. The list of current Internet- 28 Drafts is at http://datatracker.ietf.org/drafts/current/. 30 Internet-Drafts are draft documents valid for a maximum of six months 31 and may be updated, replaced, or obsoleted by other documents at any 32 time. It is inappropriate to use Internet-Drafts as reference 33 material or to cite them other than as "work in progress." 35 This Internet-Draft will expire on February 19, 2017. 37 Copyright Notice 39 Copyright (c) 2016 IETF Trust and the persons identified as the 40 document authors. All rights reserved. 42 This document is subject to BCP 78 and the IETF Trust's Legal 43 Provisions Relating to IETF Documents 44 (http://trustee.ietf.org/license-info) in effect on the date of 45 publication of this document. Please review these documents 46 carefully, as they describe your rights and restrictions with respect 47 to this document. Code Components extracted from this document must 48 include Simplified BSD License text as described in Section 4.e of 49 the Trust Legal Provisions and are provided without warranty as 50 described in the Simplified BSD License. 52 Table of Contents 54 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 55 1.1. Requirements Language . . . . . . . . . . . . . . . . . . 3 56 2. BGPsec Negotiation . . . . . . . . . . . . . . . . . . . . . 3 57 2.1. The BGPsec Capability . . . . . . . . . . . . . . . . . . 3 58 2.2. Negotiating BGPsec Support . . . . . . . . . . . . . . . 4 59 3. The BGPsec_Path Attribute . . . . . . . . . . . . . . . . . . 6 60 3.1. Secure_Path . . . . . . . . . . . . . . . . . . . . . . . 8 61 3.2. Signature_Block . . . . . . . . . . . . . . . . . . . . . 9 62 4. BGPsec Update Messages . . . . . . . . . . . . . . . . . . . 10 63 4.1. General Guidance . . . . . . . . . . . . . . . . . . . . 11 64 4.2. Constructing the BGPsec_Path Attribute . . . . . . . . . 13 65 4.3. Processing Instructions for Confederation Members . . . . 17 66 4.4. Reconstructing the AS_PATH Attribute . . . . . . . . . . 19 67 5. Processing a Received BGPsec Update . . . . . . . . . . . . . 20 68 5.1. Overview of BGPsec Validation . . . . . . . . . . . . . . 22 69 5.2. Validation Algorithm . . . . . . . . . . . . . . . . . . 23 70 6. Algorithms and Extensibility . . . . . . . . . . . . . . . . 26 71 6.1. Algorithm Suite Considerations . . . . . . . . . . . . . 26 72 6.2. Extensibility Considerations . . . . . . . . . . . . . . 27 73 7. Security Considerations . . . . . . . . . . . . . . . . . . . 28 74 7.1. Security Guarantees . . . . . . . . . . . . . . . . . . . 28 75 7.2. On the Removal of BGPsec Signatures . . . . . . . . . . . 29 76 7.3. Mitigation of Denial of Service Attacks . . . . . . . . . 30 77 7.4. Additional Security Considerations . . . . . . . . . . . 31 78 8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 31 79 9. Contributors . . . . . . . . . . . . . . . . . . . . . . . . 32 80 9.1. Authors . . . . . . . . . . . . . . . . . . . . . . . . . 32 81 9.2. Acknowledgements . . . . . . . . . . . . . . . . . . . . 33 82 10. References . . . . . . . . . . . . . . . . . . . . . . . . . 33 83 10.1. Normative References . . . . . . . . . . . . . . . . . . 33 84 10.2. Informative References . . . . . . . . . . . . . . . . . 34 85 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 35 87 1. Introduction 89 This document describes BGPsec, a mechanism for providing path 90 security for Border Gateway Protocol (BGP) [RFC4271] route 91 advertisements. That is, a BGP speaker who receives a valid BGPsec 92 update has cryptographic assurance that the advertised route has the 93 following property: Every AS on the path of ASes listed in the update 94 message has explicitly authorized the advertisement of the route to 95 the subsequent AS in the path. 97 This document specifies an optional (non-transitive) BGP path 98 attribute, BGPsec_Path. It also describes how a BGPsec-compliant BGP 99 speaker (referred to hereafter as a BGPsec speaker) can generate, 100 propagate, and validate BGP update messages containing this attribute 101 to obtain the above assurances. 103 BGPsec is intended to be used to supplement BGP Origin Validation 104 [RFC6483][RFC6811] and when used in conjunction with origin 105 validation, it is possible to prevent a wide variety of route 106 hijacking attacks against BGP. 108 BGPsec relies on the Resource Public Key Infrastructure (RPKI) 109 certificates that attest to the allocation of AS number and IP 110 address resources. (For more information on the RPKI, see RFC 6480 111 [RFC6480] and the documents referenced therein.) Any BGPsec speaker 112 who wishes to send, to external (eBGP) peers, BGP update messages 113 containing the BGPsec_Path needs to possess a private key associated 114 with an RPKI router certificate [I-D.ietf-sidr-bgpsec-pki-profiles] 115 that corresponds to the BGPsec speaker's AS number. Note, however, 116 that a BGPsec speaker does not need such a certificate in order to 117 validate received update messages containing the BGPsec_Path 118 attribute (see Section 5.2). 120 1.1. Requirements Language 122 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 123 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 124 document are to be interpreted as described in RFC 2119 [RFC2119]. 126 2. BGPsec Negotiation 128 This document defines a BGP capability [RFC5492] that allows a BGP 129 speaker to advertise to a neighbor the ability to send or to receive 130 BGPsec update messages (i.e., update messages containing the 131 BGPsec_Path attribute). 133 2.1. The BGPsec Capability 135 This capability has capability code : TBD 137 The capability length for this capability MUST be set to 3. 139 The three octets of the capability value are specified as follows. 141 BGPsec Send Capability Value 143 0 1 2 3 4 5 6 7 144 +---------------------------------------+ 145 | Version | Dir | Reserved | 146 +---------------------------------------+ 147 | | 148 +------ AFI -----+ 149 | | 150 +---------------------------------------+ 152 The first four bits of the first octet indicate the version of BGPsec 153 for which the BGP speaker is advertising support. This document 154 defines only BGPsec version 0 (all four bits set to zero). Other 155 versions of BGPsec may be defined in future documents. A BGPsec 156 speaker MAY advertise support for multiple versions of BGPsec by 157 including multiple versions of the BGPsec capability in its BGP OPEN 158 message. 160 The fifth bit of the first octet is a direction bit which indicates 161 whether the BGP speaker is advertising the capability to send BGPsec 162 update messages or receive BGPsec update messages. The BGP speaker 163 sets this bit to 0 to indicate the capability to receive BGPsec 164 update messages. The BGP speaker sets this bit to 1 to indicate the 165 capability to send BGPsec update messages. 167 The remaining three bits of the first octet are reserved for future 168 use. These bits are set to zero by the sender of the capability and 169 ignored by the receiver of the capability. 171 The second and third octets contain the 16-bit Address Family 172 Identifier (AFI) which indicates the address family for which the 173 BGPsec speaker is advertising support for BGPsec. This document only 174 specifies BGPsec for use with two address families, IPv4 and IPv6, 175 AFI values 1 and 2 respectively. BGPsec for use with other address 176 families may be specified in future documents. 178 2.2. Negotiating BGPsec Support 180 In order to indicate that a BGP speaker is willing to send BGPsec 181 update messages (for a particular address family), a BGP speaker 182 sends the BGPsec Capability (see Section 2.1) with the Direction bit 183 (the fifth bit of the first octet) set to 1. In order to indicate 184 that the speaker is willing to receive BGP update messages containing 185 the BGPsec_Path attribute (for a particular address family), a BGP 186 speaker sends the BGPsec capability with the Direction bit set to 0. 187 In order to advertise the capability to both send and receive BGPsec 188 update messages, the BGP speaker sends two copies of the BGPsec 189 capability (one with the direction bit set to 0 and one with the 190 direction bit set to 1). 192 Similarly, if a BGP speaker wishes to use BGPsec with two different 193 address families (i.e., IPv4 and IPv6) over the same BGP session, 194 then the speaker includes two instances of this capability (one for 195 each address family) in the BGP OPEN message. A BGP speaker MUST 196 support the BGP multiprotocol extension [RFC4760]. Additionally, a 197 BGP speaker MUST NOT advertise the capability of BGPsec support for a 198 particular AFI unless it has also advertised the multiprotocol 199 extension capability for the same AFI [RFC4760]. 201 In a BGPsec peering session, a peer is permitted to send update 202 messages containing the BGPsec_Path attribute if, and only if: 204 o The given peer sent the BGPsec capability for a particular version 205 of BGPsec and a particular address family with the Direction bit 206 set to 1; and 208 o The other (receiving) peer sent the BGPsec capability for the same 209 version of BGPsec and the same address family with the Direction 210 bit set to 0. 212 In such a session, we say that the use of the particular version of 213 BGPsec has been negotiated for a particular address family. BGP 214 update messages without the BGPsec_Path attribute MAY be sent within 215 a session regardless of whether or not the use of BGPsec is 216 successfully negotiated. However, if BGPsec is not successfully 217 negotiated, then BGP update messages containing the BGPsec_Path 218 attribute MUST NOT be sent. 220 This document defines the behavior of implementations in the case 221 where BGPsec version zero is the only version that has been 222 successfully negotiated. Any future document which specifies 223 additional versions of BGPsec will need to specify behavior in the 224 case that support for multiple versions is negotiated. 226 BGPsec cannot provide meaningful security guarantees without support 227 for four-byte AS numbers. Therefore, any BGP speaker that announces 228 the BGPsec capability, MUST also announce the capability for four- 229 byte AS support [RFC6793]. If a BGP speaker sends the BGPsec 230 capability but not the four-byte AS support capability then BGPsec 231 has not been successfully negotiated, and update messages containing 232 the BGPsec_Path attribute MUST NOT be sent within such a session. 234 Note that BGPsec update messages can be quite large, therefore any 235 BGPsec speaker announcing the capability to receive BGPsec messages 236 SHOULD also announce support for the capability to receive BGP 237 extended messages [I-D.ietf-idr-bgp-extended-messages]. 239 3. The BGPsec_Path Attribute 241 The BGPsec_Path attribute is an optional non-transitive BGP path 242 attribute. 244 This document registers an attribute type code for this attribute : 245 TBD 247 The BGPsec_Path attribute carries the secured information regarding 248 the path of ASes through which an update message passes. This 249 includes the digital signatures used to protect the path information. 250 We refer to those update messages that contain the BGPsec_Path 251 attribute as "BGPsec Update messages". The BGPsec_Path attribute 252 replaces the AS_PATH attribute in a BGPsec update message. That is, 253 update messages that contain the BGPsec_Path attribute MUST NOT 254 contain the AS_PATH attribute, and vice versa. 256 The BGPsec_Path attribute is made up of several parts. The following 257 high-level diagram provides an overview of the structure of the 258 BGPsec_Path attribute: 260 High-Level Diagram of the BGPsec_Path Attribute 262 +---------------------------------------------------------+ 263 | +-----------------+ | 264 | | Secure Path | | 265 | +-----------------+ | 266 | | AS X | | 267 | | pCount X | | 268 | | Flags X | | 269 | | AS Y | | 270 | | pCount Y | | 271 | | Flags Y | | 272 | | ... | | 273 | +-----------------+ | 274 | | 275 | +-----------------+ +-----------------+ | 276 | | Sig Block 1 | | Sig Block 2 | | 277 | +-----------------+ +-----------------+ | 278 | | Alg Suite 1 | | Alg Suite 2 | | 279 | | SKI X1 | | SKI X1 | | 280 | | Signature X1 | | Signature X1 | | 281 | | SKI Y1 | | SKI Y1 | | 282 | | Signature Y1 | | Signature Y1 | | 283 | | ... | | .... | | 284 | +-----------------+ +-----------------+ | 285 | | 286 +---------------------------------------------------------+ 288 The following is the specification of the format for the BGPsec_Path 289 attribute. 291 BGPsec_Path Attribute 293 +-------------------------------------------------------+ 294 | Secure_Path (variable) | 295 +-------------------------------------------------------+ 296 | Sequence of one or two Signature_Blocks (variable) | 297 +-------------------------------------------------------+ 299 The Secure_Path contains AS path information for the BGPsec update 300 message. This is logically equivalent to the information that is 301 contained in a non-BGPsec AS_PATH attribute. The information in 302 Secure_Path is used by BGPsec speakers in the same way that 303 information from the AS_PATH is used by non-BGPsec speakers. The 304 format of the Secure_Path is described below in Section 3.1. 306 The BGPsec_Path attribute will contain one or two Signature_Blocks, 307 each of which corresponds to a different algorithm suite. Each of 308 the Signature_Blocks will contain a signature segment for each AS 309 number (i.e., Secure_Path segment) in the Secure_Path. In the most 310 common case, the BGPsec_Path attribute will contain only a single 311 Signature_Block. However, in order to enable a transition from an 312 old algorithm suite to a new algorithm suite (without a flag day), it 313 will be necessary to include two Signature_Blocks (one for the old 314 algorithm suite and one for the new algorithm suite) during the 315 transition period. (See Section 6.1 for more discussion of algorithm 316 transitions.) The format of the Signature_Blocks is described below 317 in Section 3.2. 319 3.1. Secure_Path 321 Here we provide a detailed description of the Secure_Path information 322 in the BGPsec_Path attribute. 324 Secure_Path 326 +-----------------------------------------------+ 327 | Secure_Path Length (2 octets) | 328 +-----------------------------------------------+ 329 | One or More Secure_Path Segments (variable) | 330 +-----------------------------------------------+ 332 The Secure_Path Length contains the length (in octets) of the entire 333 Secure_Path (including the two octets used to express this length 334 field). As explained below, each Secure_Path segment is six octets 335 long. Note that this means the Secure_Path Length is two greater 336 than six times the number Secure_Path Segments (i.e., the number of 337 AS numbers in the path). 339 The Secure_Path contains one Secure_Path Segment for each Autonomous 340 System in the path to the originating AS of the NLRI specified in the 341 update message. 343 Secure_Path Segment 345 +----------------------------+ 346 | pCount (1 octet) | 347 +----------------------------+ 348 | Flags (1 octet) | 349 +----------------------------+ 350 | AS Number (4 octets) | 351 +----------------------------+ 353 The AS Number is the AS number of the BGP speaker that added this 354 Secure_Path segment to the BGPsec_Path attribute. (See Section 4 for 355 more information on populating this field.) 356 The pCount field contains the number of repetitions of the associated 357 autonomous system number that the signature covers. This field 358 enables a BGPsec speaker to mimic the semantics of prepending 359 multiple copies of their AS to the AS_PATH without requiring the 360 speaker to generate multiple signatures. The pCount field is also 361 useful in managing route servers (see Section 4.2) and AS Number 362 migrations, see [I-D.ietf-sidr-as-migration] for details. 364 The first bit of the Flags field is the Confed_Segment flag. The 365 Confed_Segment flag is set to one to indicate that the BGPsec speaker 366 that constructed this Secure_Path segment is sending the update 367 message to a peer AS within the same Autonomous System confederation 368 [RFC5065]. (That is, the Confed_Segment flag is set in a BGPsec 369 update message whenever, in a non-BGPsec update message, the BGP 370 speaker's AS would appear in a AS_PATH segment of type 371 AS_CONFED_SEQUENCE.) In all other cases the Confed_Segment flag is 372 set to zero. 374 The remaining seven bits of the Flags MUST be set to zero by the 375 sender, and ignored by the receiver. Note, however, that the 376 signature is computed over all eight bits of the flags field. 378 3.2. Signature_Block 380 Here we provide a detailed description of the Signature_Blocks in the 381 BGPsec_Path attribute. 383 Signature_Block 385 +---------------------------------------------+ 386 | Signature_Block Length (2 octets) | 387 +---------------------------------------------+ 388 | Algorithm Suite Identifier (1 octet) | 389 +---------------------------------------------+ 390 | Sequence of Signature Segments (variable) | 391 +---------------------------------------------+ 393 The Signature_Block Length is the total number of octets in the 394 Signature_Block (including the two octets used to express this length 395 field). 397 The Algorithm Suite Identifier is a one-octet identifier specifying 398 the digest algorithm and digital signature algorithm used to produce 399 the digital signature in each Signature Segment. An IANA registry of 400 algorithm identifiers for use in BGPsec is specified in the BGPsec 401 algorithms document [I-D.ietf-sidr-bgpsec-algs]. 403 A Signature_Block has exactly one Signature Segment for each 404 Secure_Path Segment in the Secure_Path portion of the BGPsec_Path 405 Attribute. (That is, one Signature Segment for each distinct AS on 406 the path for the NLRI in the Update message.) 408 Signature Segments 410 +---------------------------------------------+ 411 | Subject Key Identifier (20 octets) | 412 +---------------------------------------------+ 413 | Signature Length (2 octets) | 414 +---------------------------------------------+ 415 | Signature (variable) | 416 +---------------------------------------------+ 418 The Subject Key Identifier contains the value in the Subject Key 419 Identifier extension of the RPKI router certificate 420 [I-D.ietf-sidr-bgpsec-pki-profiles] that is used to verify the 421 signature (see Section 5 for details on validity of BGPsec update 422 messages). 424 The Signature Length field contains the size (in octets) of the value 425 in the Signature field of the Signature Segment. 427 The Signature contains a digital signature that protects the NLRI and 428 the BGPsec_Path attribute (see Section 4 and Section 5 for details on 429 signature generation and validation, respectively). 431 4. BGPsec Update Messages 433 Section 4.1 provides general guidance on the creation of BGPsec 434 Update Messages -- that is, update messages containing the 435 BGPsec_Path attribute. 437 Section 4.2 specifies how a BGPsec speaker generates the BGPsec_Path 438 attribute to include in a BGPsec Update message. 440 Section 4.3 contains special processing instructions for members of 441 an autonomous system confederation [RFC5065]. A BGPsec speaker that 442 is not a member of such a confederation MUST set the Flags field of 443 the Secure_Path Segment to zero in all BGPsec update messages it 444 sends. 446 Section 4.4 contains instructions for reconstructing the AS_PATH 447 attribute in cases where a BGPsec speaker receives an update message 448 with a BGPsec_Path attribute and wishes to propagate the update 449 message to a peer who does not support BGPsec. 451 4.1. General Guidance 453 The information protected by the signature on a BGPsec update message 454 includes the AS number of the peer to whom the update message is 455 being sent. Therefore, if a BGPsec speaker wishes to send a BGPsec 456 update to multiple BGP peers, it must generate a separate BGPsec 457 update message for each unique peer AS to whom the update message is 458 sent. 460 A BGPsec update message MUST advertise a route to only a single NLRI. 461 This is because a BGPsec speaker receiving an update message with 462 multiple NLRI would be unable to construct a valid BGPsec update 463 message (i.e., valid path signatures) containing a subset of the NLRI 464 in the received update. If a BGPsec speaker wishes to advertise 465 routes to multiple NLRI, then it MUST generate a separate BGPsec 466 update message for each NLRI. Additionally, a BGPsec update message 467 MUST use the MP_REACH_NLRI [RFC4760] attribute to encode the NLRI. 469 The BGPsec_Path attribute and the AS_PATH attribute are mutually 470 exclusive. That is, any update message containing the BGPsec_Path 471 attribute MUST NOT contain the AS_PATH attribute. The information 472 that would be contained in the AS_PATH attribute is instead conveyed 473 in the Secure_Path portion of the BGPsec_Path attribute. 475 In order to create or add a new signature to a BGPsec update message 476 with a given algorithm suite, the BGPsec speaker must possess a 477 private key suitable for generating signatures for this algorithm 478 suite. Additionally, this private key must correspond to the public 479 key in a valid Resource PKI end-entity certificate whose AS number 480 resource extension includes the BGPsec speaker's AS number 481 [I-D.ietf-sidr-bgpsec-pki-profiles]. Note also that new signatures 482 are only added to a BGPsec update message when a BGPsec speaker is 483 generating an update message to send to an external peer (i.e., when 484 the AS number of the peer is not equal to the BGPsec speaker's own AS 485 number). Therefore, a BGPsec speaker who only sends BGPsec update 486 messages to peers within its own AS does not need to possess any 487 private signature keys. 489 The Resource PKI enables the legitimate holder of IP address 490 prefix(es) to issue a signed object, called a Route Origination 491 Authorization (ROA), that authorizes a given AS to originate routes 492 to a given set of prefixes (see RFC 6482 [RFC6482]). It is expected 493 that most relying parties will utilize BGPsec in tandem with origin 494 validation (see RFC 6483 [RFC6483] and RFC 6811 [RFC6811]). 495 Therefore, it is RECOMMENDED that a BGPsec speaker only originate a 496 BGPsec update advertising a route for a given prefix if there exists 497 a valid ROA authorizing the BGPsec speaker's AS to originate routes 498 to this prefix. 500 If a BGPsec router has received only a non-BGPsec update message 501 (without the BGPsec_Path attribute), containing the AS_PATH 502 attribute, from a peer for a given prefix then it MUST NOT attach a 503 BGPsec_Path attribute when it propagates the update message. (Note 504 that a BGPsec router may also receive a non-BGPsec update message 505 from an internal peer without the AS_PATH attribute, i.e., with just 506 the NLRI in it. In that case, the prefix is originating from that 507 AS, and if it is selected for advertisement, the BGPsec speaker 508 SHOULD attach a BGPsec_Path attribute and send a signed route (for 509 that prefix) to its external BGPsec-speaking peers.) 511 Conversely, if a BGPsec router has received a BGPsec update message 512 (with the BGPsec_Path attribute) from a peer for a given prefix and 513 it chooses to propagate that peer's route for the prefix, then it 514 SHOULD propagate the route as a BGPsec update message containing the 515 BGPsec_Path attribute. 517 Note that removing BGPsec signatures (i.e., propagating a route 518 advertisement without the BGPsec_Path attribute) has significant 519 security ramifications. (See Section 7 for discussion of the 520 security ramifications of removing BGPsec signatures.) Therefore, 521 when a route advertisement is received via a BGPsec update message, 522 propagating the route advertisement without the BGPsec_Path attribute 523 is NOT RECOMMENDED, unless the message is sent to a peer that did not 524 advertise the capability to receive BGPsec update messages (see 525 Section 4.4). 527 Furthermore, note that when a BGPsec speaker propagates a route 528 advertisement with the BGPsec_Path attribute it is not attesting to 529 the validation state of the update message it received. (See 530 Section 7 for more discussion of the security semantics of BGPsec 531 signatures.) 533 If the BGPsec speaker is producing an update message which would, in 534 the absence of BGPsec, contain an AS_SET (e.g., the BGPsec speaker is 535 performing proxy aggregation), then the BGPsec speaker MUST NOT 536 include the BGPsec_Path attribute. In such a case, the BGPsec 537 speaker must remove any existing BGPsec_Path in the received 538 advertisement(s) for this prefix and produce a traditional (non- 539 BGPsec) update message. It should be noted that BCP 172 [RFC6472] 540 recommends against the use of AS_SET and AS_CONFED_SET in the AS_PATH 541 of BGP updates. 543 The case where the BGPsec speaker sends a BGPsec update message to an 544 internal (iBGP) peer is quite simple. When originating a new route 545 advertisement and sending it to an internal peer, the BGPsec speaker 546 omits the BGPsec_Path attribute. When propagating a received route 547 advertisement to an internal peer, the BGPsec speaker typically 548 populates the BGPsec_Path attribute by copying the BGPsec_Path 549 attribute from the received update message. That is, the BGPsec_Path 550 attribute is copied verbatim. However, in the case that the BGPsec 551 speaker is performing an AS Migration, the BGPsec speaker may add an 552 additional signature on ingress before copying the BGPsec_Path 553 attribute (see [I-D.ietf-sidr-as-migration] for more details). Note 554 that when a BGPsec speaker chooses to forward a BGPsec update message 555 to an iBGP peer, the BGPsec attribute SHOULD NOT be removed, unless 556 the peer doesn't support BGPsec. In particular, the BGPsec attribute 557 SHOULD NOT be removed even in the case where the BGPsec update 558 message has not been successfully validated. (See Section 5 for more 559 information on validation, and Section 7 for the security 560 ramifications of removing BGPsec signatures.) 562 4.2. Constructing the BGPsec_Path Attribute 564 When a BGPsec speaker receives a BGPsec update message containing a 565 BGPsec_Path attribute (with one or more signatures) from an (internal 566 or external) peer, it may choose to propagate the route advertisement 567 by sending it to its other (internal or external) peers. When 568 sending said route advertisement to an internal BGPsec-speaking peer, 569 the BGPsec_Path attribute SHALL NOT be modified. When sending said 570 route advertisement to an external BGPsec-speaking peer, the 571 following procedures are used to form or update the BGPsec_Path 572 attribute. 574 To generate the BGPsec_Path attribute on the outgoing update message, 575 the BGPsec speaker first generates a new Secure_Path Segment. Note 576 that if the BGPsec speaker is not the origin AS and there is an 577 existing BGPsec_Path attribute, then the BGPsec speaker prepends its 578 new Secure_Path Segment (places in first position) onto the existing 579 Secure_Path. 581 The AS number in this Secure_Path segment MUST match the AS number in 582 the AS number resource extension field of the Resource PKI router 583 certificate(s) that will be used to verify the digital signature(s) 584 constructed by this BGPsec speaker 585 [I-D.ietf-sidr-bgpsec-pki-profiles]. 587 The pCount field of the Secure_Path Segment is typically set to the 588 value 1. However, a BGPsec speaker may set the pCount field to a 589 value greater than 1. Setting the pCount field to a value greater 590 than one has the same semantics as repeating an AS number multiple 591 times in the AS_PATH of a non-BGPsec update message (e.g., for 592 traffic engineering purposes). 594 To prevent unnecessary processing load in the validation of BGPsec 595 signatures, a BGPsec speaker SHOULD NOT produce multiple consecutive 596 Secure_Path Segments with the same AS number. This means that to 597 achieve the semantics of prepending the same AS number k times, a 598 BGPsec speaker SHOULD produce a single Secure_Path Segment -- with 599 pCount of k -- and a single corresponding Signature Segment. 601 A route server that participates in the BGP control plane, but does 602 not act as a transit AS in the data plane, may choose to set pCount 603 to 0. This option enables the route server to participate in BGPsec 604 and obtain the associated security guarantees without increasing the 605 effective length of the AS path. (Note that BGPsec speakers compute 606 the effective length of the AS path by summing the pCount values in 607 the BGPsec_Path attribute, see Section 5.) However, when a route 608 server sets the pCount value to 0, it still inserts its AS number 609 into the Secure_Path segment, as this information is needed to 610 validate the signature added by the route server. (See 611 [I-D.ietf-sidr-as-migration] for a discussion of setting pCount to 0 612 to facilitate AS Number Migration.) BGPsec speakers SHOULD drop 613 incoming update messages with pCount set to zero in cases where the 614 BGPsec speaker does not expect its peer to set pCount to zero. (That 615 is, pCount is only to be set to zero in cases such as route servers 616 or AS Number Migration where the BGPsec speaker's peer expects pCount 617 to be set to zero.) 619 Next, the BGPsec speaker generates one or two Signature_Blocks. 620 Typically, a BGPsec speaker will use only a single algorithm suite, 621 and thus create only a single Signature_Block in the BGPsec_Path 622 attribute. However, to ensure backwards compatibility during a 623 period of transition from a 'current' algorithm suite to a 'new' 624 algorithm suite, it will be necessary to originate update messages 625 that contain a Signature_Block for both the 'current' and the 'new' 626 algorithm suites (see Section 6.1). 628 If the received BGPsec update message contains two Signature_Blocks 629 and the BGPsec speaker supports both of the corresponding algorithm 630 suites, then the new update message generated by the BGPsec speaker 631 SHOULD include both of the Signature_Blocks. If the received BGPsec 632 update message contains two Signature_Blocks and the BGPsec speaker 633 only supports one of the two corresponding algorithm suites, then the 634 BGPsec speaker MUST remove the Signature_Block corresponding to the 635 algorithm suite that it does not understand. If the BGPsec speaker 636 does not support the algorithm suites in any of the Signature_Blocks 637 contained in the received update message, then the BGPsec speaker 638 MUST NOT propagate the route advertisement with the BGPsec_Path 639 attribute. (That is, if it chooses to propagate this route 640 advertisement at all, it must do so as an unsigned BGP update 641 message. See Section 4.4 for more information on converting to an 642 unsigned BGP message.) 643 Note that in the case where the BGPsec_Path has two Signature_Blocks 644 (corresponding to different algorithm suites), the validation 645 algorithm (see Section 5.2) deems a BGPsec update message to be 646 'Valid' if there is at least one supported algorithm suite (and 647 corresponding Signature_Block) that is deemed 'Valid'. This means 648 that a 'Valid' BGPsec update message may contain a Signature_Block 649 which is not deemed 'Valid' (e.g., contains signatures that BGPsec 650 does not successfully verify). Nonetheless, such Signature_Blocks 651 MUST NOT be removed. (See Section 7 for a discussion of the security 652 ramifications of this design choice.) 654 For each Signature_Block corresponding to an algorithm suite that the 655 BGPsec speaker does support, the BGPsec speaker adds a new Signature 656 Segment to the Signature_Block. This Signature Segment is prepended 657 to the list of Signature Segments (placed in the first position) so 658 that the list of Signature Segments appear in the same order as the 659 corresponding Secure_Path segments. The BGPsec speaker populates the 660 fields of this new signature segment as follows. 662 The Subject Key Identifier field in the new segment is populated with 663 the identifier contained in the Subject Key Identifier extension of 664 the RPKI router certificate corresponding to the BGPsec speaker 665 [I-D.ietf-sidr-bgpsec-pki-profiles]. This Subject Key Identifier 666 will be used by recipients of the route advertisement to identify the 667 proper certificate to use in verifying the signature. 669 The Signature field in the new segment contains a digital signature 670 that binds the NLRI and BGPsec_Path attribute to the RPKI router 671 certificate corresponding to the BGPsec speaker. The digital 672 signature is computed as follows: 674 o For clarity, let us number the Secure_Path and corresponding 675 Signature Segments from 1 to N as follows. Let Secure_Path 676 Segment 1 and Signature Segment 1 be the segments produced by the 677 origin AS. Let Secure_Path Segment 2 and Signature Segment 2 be 678 the segments added by the next AS after the origin. Continue this 679 method of numbering and ultimately let Secure_Path Segment N be 680 the Secure_Path segment that is being added by the current AS. 682 o In order to construct the digital signature for Signature Segment 683 N (the signature segment being produced by the current AS), first 684 construct the following sequence of octets to be hashed. 686 Sequence of Octets to be Hashed 688 +------------------------------------+ 689 | Target AS Number | 690 +------------------------------------+ -\ 691 | Signature Segment : N-1 | \ 692 +------------------------------------+ | 693 | Secure_Path Segment : N | | 694 +------------------------------------+ \ 695 ... > For N Hops 696 +------------------------------------+ / 697 | Signature Segment : 1 | | 698 +------------------------------------+ | 699 | Secure_Path Segment : 2 | / 700 +------------------------------------+ -/ 701 | Secure_Path Segment : 1 | 702 +------------------------------------+ 703 | Algorithm Suite Identifier | 704 +------------------------------------+ 705 | AFI | 706 +------------------------------------+ 707 | SAFI | 708 +------------------------------------+ 709 | NLRI | 710 +------------------------------------+ 712 In this sequence, the Target AS Number is the AS to whom the 713 BGPsec speaker intends to send the update message. (Note that the 714 Target AS number is the AS number announced by the peer in the 715 OPEN message of the BGP session within which the update is sent.) 716 The Secure_Path and Signature Segments (1 through N-1) are 717 obtained from the BGPsec_Path attribute. Finally, the Address 718 Family Identifier (AFI), Subsequent Address Family Identifier 719 (SAFI), and Network Layer Reachability Information (NLRI) fields 720 are obtained from the MP_REACH_NLRI attribute. Additionally, in 721 the Prefix field of the NLRI (from MP_REACH_NLRI), all of the 722 trailing bits MUST be set to zero when constructing this sequence. 724 o Apply to this octet sequence the digest algorithm (for the 725 algorithm suite of this Signature_Block) to obtain a digest value. 727 o Apply to this digest value the signature algorithm, (for the 728 algorithm suite of this Signature_Block) to obtain the digital 729 signature. Then populate the Signature Field with this digital 730 signature. 732 The Signature Length field is populated with the length (in octets) 733 of the value in the Signature field. 735 4.3. Processing Instructions for Confederation Members 737 Members of autonomous system confederations [RFC5065] MUST 738 additionally follow the instructions in this section for processing 739 BGPsec update messages. 741 When a confederation member sends a BGPsec update message to a peer 742 that is a member of the same Member-AS, the confederation member 743 SHALL NOT modify the BGPsec_Path attribute. When a confederation 744 member sends a BGPsec update message to a peer that is a member of 745 the same confederation but is a different Member-AS, the 746 confederation member puts its (private) Member-AS Number (as opposed 747 to the public AS Confederation Identifier) in the AS Number field of 748 the Secure_Path Segment that it adds to the BGPsec update message. 749 Additionally, in this case, the confederation member that generates 750 the Secure_Path Segment sets the Confed_Segment flag to one. This 751 means that in a BGPsec update message, an AS number appears in a 752 Secure_Path Segment with the Confed_Segment flag set whenever, in a 753 non-BGPsec update message, the AS number would appear in a segment of 754 type AS_CONFED_SEQUENCE. 756 Within a confederation, the verification of BGPsec signatures added 757 by other members of the confederation is optional. If a 758 confederation chooses not to have its members verify signatures added 759 by other confederation members, then when sending a BGPsec update 760 message to a peer that is a member of the same confederation, the 761 confederation members MAY set the Signature field within the 762 Signature Segment that it generates to be zero (in lieu of 763 calculating the correct digital signature as described in 764 Section 4.2). Note that if a confederation chooses not to verify 765 digital signatures within the confederation, then BGPsec is able to 766 provide no assurances about the integrity of the (private) Member-AS 767 Numbers placed in Secure_Path segments where the Confed_Segment flag 768 is set to one. 770 When a confederation member receives a BGPsec update message from a 771 peer within the confederation and propagates it to a peer outside the 772 confederation, it needs to remove all of the Secure_Path Segments 773 added by confederation members as well as the corresponding Signature 774 Segments. To do this, the confederation member propagating the route 775 outside the confederation does the following: 777 o First, starting with the most recently added Secure_Path segment, 778 remove all of the consecutive Secure_Path segments that have the 779 Confed_Segment flag set to one. Stop this process once a 780 Secure_Path segment is reached which has its Confed_Segment flag 781 set to zero. Keep a count of the number of segments removed in 782 this fashion. 784 o Second, starting with the most recently added Signature Segment, 785 remove a number of Signature Segments equal to the number of 786 Secure_Path Segments removed in the previous step. (That is, 787 remove the K most recently added signature segments, where K is 788 the number of Secure_Path Segments removed in the previous step.) 790 o Finally, add a Secure_Path Segment containing, in the AS field, 791 the AS Confederation Identifier (the public AS number of the 792 confederation) as well as a corresponding Signature Segment. Note 793 that all fields other that the AS field are populated as per 794 Section 4.2. 796 When validating a received BGPsec update message, confederation 797 members need to make the following adjustment to the algorithm 798 presented in Section 5.2. When a confederation member processes 799 (validates) a Signature Segment and its corresponding Secure_Path 800 Segment, the confederation member must note the following. For a 801 signature produced by a peer BGPsec speaker outside of a 802 confederation, the Target AS will always be the AS Confederation 803 Identifier (the public AS number of the confederation) as opposed to 804 the Member-AS Number. 806 To handle this case, when a BGPsec speaker (that is a confederation 807 member) processes a current Secure_Path Segment that has the 808 Confed_Segment flag set to zero, if the next most recently added 809 Secure_Path segment has the Confed_Segment flag set to one then, when 810 computing the digest for the current Secure_Path segment, the BGPsec 811 speaker takes the Target AS Number to be the AS Confederation 812 Identifier of the validating BGPsec speaker's own confederation. 813 (Note that the algorithm in Section 5.2 processes Secure_Path 814 Segments in order from most recently added to least recently added, 815 therefore this special case will apply to the first Secure_Path 816 segment that the algorithm encounters that has the Confed_Segment 817 flag set to zero.) 819 Finally, as discussed above, an AS confederation may optionally 820 decide that its members will not verify digital signatures added by 821 members. In such a federation, when a confederation member runs the 822 algorithm in Section 5.2, the confederation member, during processing 823 of a Signature Segment, first checks whether the Confed_Sequence flag 824 in the corresponding Secure_Path segment is set to one. If the 825 Confed_Sequence flag is set to one in the corresponding Secure_Path 826 segment, the confederation member does not perform any further checks 827 on the Signature Segment and immediately moves on to the next 828 Signature Segment (and checks its corresponding Secure_Path segment). 829 Note that as specified in Section 5.2, it is an error when a BGPsec 830 speaker receives from a peer, who is not in the same AS 831 confederation, a BGPsec update containing a Confed_Sequence flag set 832 to one. (As discussed in Section 5.2, any error in the BGPsec_Path 833 attribute MUST be handled using the "treat-as-withdraw", approach as 834 defined in RFC 7606 [RFC7606].) 836 4.4. Reconstructing the AS_PATH Attribute 838 BGPsec update messages do not contain the AS_PATH attribute. 839 However, the AS_PATH attribute can be reconstructed from the 840 BGPsec_Path attribute. This is necessary in the case where a route 841 advertisement is received via a BGPsec update message and then 842 propagated to a peer via a non-BGPsec update message (e.g., because 843 the latter peer does not support BGPsec). Note that there may be 844 additional cases where an implementation finds it useful to perform 845 this reconstruction. Before attempting to reconstruct an AS_PATH for 846 the purpose of forwarding an unsigned (non-BGPsec) update to a peer, 847 a BGPsec speaker MUST perform the basic integrity checks listed in 848 Section 5.2 to ensure that the received BGPsec update is properly 849 formed. 851 The AS_PATH attribute can be constructed from the BGPsec_Path 852 attribute as follows. Starting with an empty AS_PATH attribute, 853 process the Secure_Path segments in order from least-recently added 854 (corresponding to the origin) to most-recently added. For each 855 Secure_Path segment perform the following steps: 857 1. If the Confed_Segment flag in the Secure_Path segment is set to 858 one, then look at the most-recently added segment in the AS_PATH. 860 * In the case where the AS_PATH is empty or in the case where 861 the most-recently added segment is of type AS_SEQUENCE then 862 add (prepend to the AS_PATH) a new AS_PATH segment of type 863 AS_CONFED_SEQUENCE. This segment of type AS_CONFED_SEQUENCE 864 shall contain a number of elements equal to the pCount field 865 in the current Secure_Path segment. Each of these elements 866 shall be the AS number contained in the current Secure_Path 867 segment. (That is, if the pCount field is X, then the segment 868 of type AS_CONFED_SEQUENCE contains X copies of the 869 Secure_Path segment's AS Number field.) 871 * In the case where the most-recently added segment in the 872 AS_PATH is of type AS_CONFED_SEQUENCE then add (prepend to the 873 segment) a number of elements equal to the pCount field in the 874 current Secure_Path segment. The value of each of these 875 elements shall be the AS number contained in the current 876 Secure_Path segment. (That is, if the pCount field is X, then 877 add X copies of the Secure_Path segment's AS Number field to 878 the existing AS_CONFED_SEQUENCE.) 880 2. If the Confed_Segment flag in the Secure_Path segment is set to 881 zero, then look at the most-recently added segment in the 882 AS_PATH. 884 * In the case where the AS_PATH is empty, and the pCount field 885 in the Secure_Path segment is greater than zero, add (prepend 886 to the AS_PATH) a new AS_PATH segment of type AS_SEQUENCE. 887 This segment of type AS_SEQUENCE shall contain a number of 888 elements equal to the pCount field in the current Secure_Path 889 segment. Each of these elements shall be the AS number 890 contained in the current Secure_Path segment. (That is, if 891 the pCount field is X, then the segment of type AS_SEQUENCE 892 contains X copies of the Secure_Path segment's AS Number 893 field.) 895 * In the case where the most recently added segment in the 896 AS_PATH is of type AS_SEQUENCE then add (prepend to the 897 segment) a number of elements equal to the pCount field in the 898 current Secure_Path segment. The value of each of these 899 elements shall be the AS number contained in the current 900 Secure_Path segment. (That is, if the pCount field is X, then 901 add X copies of the Secure_Path segment's AS Number field to 902 the existing AS_SEQUENCE.) 904 As part of the above described procedure, the following additional 905 actions are performed in order not to exceed the size limitations of 906 AS_SEQUENCE and AS_CONFED_SEQUENCE. While adding the next 907 Secure_Path segment (with its prepends, if any) to the AS_PATH being 908 assembled, if it would cause the AS_SEQUENCE (or AS_CONFED_SEQUENCE) 909 at hand to exceed the 255 ASN per segment limit [RFC4271][RFC5065], 910 then the BGPsec speaker would follow the recommendations in RFC 4271 911 [RFC4271] and RFC 5065 [RFC5065] of creating another segment of the 912 same type (AS_SEQUENCE or AS_CONFED_SEQUENCE) and continue filling 913 that. 915 5. Processing a Received BGPsec Update 917 Upon receiving a BGPsec update message from an external (eBGP) peer, 918 a BGPsec speaker SHOULD validate the message to determine the 919 authenticity of the path information contained in the BGPsec_Path 920 attribute. Typically, a BGPsec speaker will also wish to perform 921 origin validation (see RFC 6483 [RFC6483] and RFC 6811 [RFC6811]) on 922 an incoming BGPsec update message, but such validation is independent 923 of the validation described in this section. 925 Section 5.1 provides an overview of BGPsec validation and Section 5.2 926 provides a specific algorithm for performing such validation. (Note 927 that an implementation need not follow the specific algorithm in 928 Section 5.2 as long as the input/output behavior of the validation is 929 identical to that of the algorithm in Section 5.2.) During 930 exceptional conditions (e.g., the BGPsec speaker receives an 931 incredibly large number of update messages at once) a BGPsec speaker 932 MAY temporarily defer validation of incoming BGPsec update messages. 933 The treatment of such BGPsec update messages, whose validation has 934 been deferred, is a matter of local policy. However, an 935 implementation SHOULD ensure that deferment of validation and status 936 of deferred messages is visible to the operator. 938 The validity of BGPsec update messages is a function of the current 939 RPKI state. When a BGPsec speaker learns that RPKI state has changed 940 (e.g., from an RPKI validating cache via the RPKI-to-Router protocol 941 [I-D.ietf-sidr-rpki-rtr-rfc6810-bis]), the BGPsec speaker MUST re-run 942 validation on all affected update messages stored in its Adj-RIB-In. 943 For example, when a given RPKI certificate ceases to be valid (e.g., 944 it expires or is revoked), all update messages containing a signature 945 whose SKI matches the SKI in the given certificate must be re- 946 assessed to determine if they are still valid. If this reassessment 947 determines that the validity state of an update has changed then, 948 depending on local policy, it may be necessary to re-run best path 949 selection. 951 BGPsec update messages do not contain an AS_PATH attribute. 952 Therefore, a BGPsec speaker MUST utilize the AS path information in 953 the BGPsec_Path attribute in all cases where it would otherwise use 954 the AS path information in the AS_PATH attribute. The only exception 955 to this rule is when AS path information must be updated in order to 956 propagate a route to a peer (in which case the BGPsec speaker follows 957 the instructions in Section 4). Section 4.4 provides an algorithm 958 for constructing an AS_PATH attribute from a BGPsec_Path attribute. 959 Whenever the use of AS path information is called for (e.g., loop 960 detection, or use of AS path length in best path selection) the 961 externally visible behavior of the implementation shall be the same 962 as if the implementation had run the algorithm in Section 4.4 and 963 used the resulting AS_PATH attribute as it would for a non-BGPsec 964 update message. 966 Many signature algorithms are non-deterministic. That is, many 967 signature algorithms will produce different signatures each time they 968 are run (even when they are signing the same data with the same key). 969 Therefore, if an implementation receives a BGPsec update from a peer 970 and later receives a second BGPsec update message from the same peer, 971 the implementation SHOULD treat the second message as a duplicate 972 update message if it differs from the first update message only in 973 the Signature fields (within the BGPsec_Path attribute). That is, if 974 all the fields in the second update are identical to the fields in 975 the first update message, except for the Signature fields, then the 976 second update message should be treated as a duplicate of the first 977 update message. Note that if other fields (e.g., the Subject Key 978 Identifier field) within a Signature segment differ between two 979 update messages then the two updates are not duplicates. 981 With regards to the processing of duplicate update messages, if the 982 first update message is valid, then an implementation SHOULD NOT run 983 the validation procedure on the second, duplicate update message 984 (even if the bits of the signature field are different). If the 985 first update message is not valid, then an implementation SHOULD run 986 the validation procedure on the second duplicate update message (as 987 the signatures in the second update may be valid even though the 988 first contained a signature that was invalid). 990 5.1. Overview of BGPsec Validation 992 Validation of a BGPsec update messages makes use of data from RPKI 993 certificates. In particular, it is necessary that the recipient have 994 access to the following data obtained from valid RPKI certificates: 995 the AS Number, Public Key and Subject Key Identifier from each valid 996 RPKI router certificate. 998 Note that the BGPsec speaker could perform the validation of RPKI 999 certificates on its own and extract the required data, or it could 1000 receive the same data from a trusted cache that performs RPKI 1001 validation on behalf of (some set of) BGPsec speakers. (For example, 1002 the trusted cache could deliver the necessary validity information to 1003 the BGPsec speaker using the router key PDU 1004 [I-D.ietf-sidr-rtr-keying] for the RPKI-to-Router protocol 1005 [I-D.ietf-sidr-rpki-rtr-rfc6810-bis].) 1007 To validate a BGPsec update message containing the BGPsec_Path 1008 attribute, the recipient performs the validation steps specified in 1009 Section 5.2. The validation procedure results in one of two states: 1010 'Valid' and 'Not Valid'. 1012 It is expected that the output of the validation procedure will be 1013 used as an input to BGP route selection. That said, BGP route 1014 selection, and thus the handling of the validation states is a matter 1015 of local policy, and is handled using local policy mechanisms. 1016 Implementations SHOULD enable operators to set such local policy on a 1017 per-session basis. (That is, we expect some operators will choose to 1018 treat BGPsec validation status differently for update messages 1019 received over different BGP sessions.) 1021 It is expected that BGP peers will generally prefer routes received 1022 via 'Valid' BGPsec update messages over both routes received via 'Not 1023 Valid' BGPsec update messages and routes received via update messages 1024 that do not contain the BGPsec_Path attribute. However, BGPsec 1025 specifies no changes to the BGP decision process. (See 1026 [I-D.ietf-sidr-bgpsec-ops] for related operational considerations.) 1028 BGPsec validation needs only be performed at the eBGP edge. The 1029 validation status of a BGP signed/unsigned update MAY be conveyed via 1030 iBGP from an ingress edge router to an egress edge router via some 1031 mechanism, according to local policy within an AS. As discussed in 1032 Section 4, when a BGPsec speaker chooses to forward a (syntactically 1033 correct) BGPsec update message, it SHOULD be forwarded with its 1034 BGPsec_Path attribute intact (regardless of the validation state of 1035 the update message). Based entirely on local policy, an egress 1036 router receiving a BGPsec update message from within its own AS MAY 1037 choose to perform its own validation. 1039 5.2. Validation Algorithm 1041 This section specifies an algorithm for validation of BGPsec update 1042 messages. A conformant implementation MUST include a BGPsec update 1043 validation algorithm that is functionally equivalent to the 1044 externally visible behavior of this algorithm. 1046 First, the recipient of a BGPsec update message performs a check to 1047 ensure that the message is properly formed. Specifically, the 1048 recipient performs the following checks: 1050 1. Check to ensure that the entire BGPsec_Path attribute is 1051 syntactically correct (conforms to the specification in this 1052 document). 1054 2. Check that each Signature_Block contains one Signature segment 1055 for each Secure_Path segment in the Secure_Path portion of the 1056 BGPsec_Path attribute. (Note that the entirety of each 1057 Signature_Block must be checked to ensure that it is well formed, 1058 even though the validation process may terminate before all 1059 signatures are cryptographically verified.) 1061 3. Check that the update message does not contain an AS_PATH 1062 attribute. 1064 4. If the update message was received from a peer that is not a 1065 member of the BGPsec speaker's AS confederation, check to ensure 1066 that none of the Secure_Path segments contain a Flags field with 1067 the Confed_Sequence flag set to one. 1069 5. If the update message was received from a peer that is not 1070 expected to set pCount equal to zero (see Section 4.2) then check 1071 to ensure that the pCount field in the most-recently added 1072 Secure_Path segment is not equal to zero. 1074 If any of these checks fail, it is an error in the BGPsec_Path 1075 attribute. Any of these errors in the BGPsec_Path attribute are 1076 handled as per RFC 7606 [RFC7606]. BGPsec speakers MUST handle these 1077 errors using the "treat-as-withdraw" approach as defined in RFC 7606 1078 [RFC7606]. 1080 Next, the BGPsec speaker examines the Signature_Blocks in the 1081 BGPsec_Path attribute. A Signature_Block corresponding to an 1082 algorithm suite that the BGPsec speaker does not support is not 1083 considered in validation. If there is no Signature_Block 1084 corresponding to an algorithm suite that the BGPsec speaker supports, 1085 then the BGPsec speaker MUST treat the update message in the same 1086 manner that the BGPsec speaker would treat an (unsigned) update 1087 message that arrived without a BGPsec_Path attribute. 1089 For each remaining Signature_Block (corresponding to an algorithm 1090 suite supported by the BGPsec speaker), the BGPsec speaker iterates 1091 through the Signature segments in the Signature_Block, starting with 1092 the most recently added segment (and concluding with the least 1093 recently added segment). Note that there is a one-to-one 1094 correspondence between Signature segments and Secure_Path segments 1095 within the BGPsec_Path attribute. The following steps make use of 1096 this correspondence. 1098 o (Step 0): For clarity, let us number the Secure_Path and 1099 corresponding Signature Segments from 1 to N as follows. Let 1100 Secure_Path Segment 1 and Signature Segment 1 be the segments 1101 produced by the origin AS. Let Secure_Path Segment 2 and 1102 Signature Segment 2 be the segments added by the next AS after the 1103 origin. Continue this method of numbering and ultimately let 1104 Signature Segment N be the Signature Segment that is currently 1105 being verified and let Secure_Path Segment N be the corresponding 1106 Secure_Path Segment. 1108 o (Step I): Locate the public key needed to verify the signature (in 1109 the current Signature segment). To do this, consult the valid 1110 RPKI router certificate data and look up all valid (AS, SKI, 1111 Public Key) triples in which the AS matches the AS number in the 1112 corresponding Secure_Path segment. Of these triples that match 1113 the AS number, check whether there is an SKI that matches the 1114 value in the Subject Key Identifier field of the Signature 1115 segment. If this check finds no such matching SKI value, then 1116 mark the entire Signature_Block as 'Not Valid' and proceed to the 1117 next Signature_Block. 1119 o (Step II): Compute the digest function (for the given algorithm 1120 suite) on the appropriate data. 1122 In order to verify the digital signature in Signature Segment N, 1123 construct the following sequence of octets to be hashed. 1125 Sequence of Octets to be Hashed for Signature Verification 1127 +------------------------------------+ 1128 | Target AS Number | 1129 +------------------------------------+ -\ 1130 | Signature Segment : N-1 | \ 1131 +------------------------------------+ | 1132 | Secure_Path Segment : N | | 1133 +------------------------------------+ \ 1134 ... > For N Hops 1135 +------------------------------------+ / 1136 | Signature Segment : 1 | | 1137 +------------------------------------+ | 1138 | Secure_Path Segment : 2 | / 1139 +------------------------------------+ -/ 1140 | Secure_Path Segment : 1 | 1141 +------------------------------------+ 1142 | Algorithm Suite Identifier | 1143 +------------------------------------+ 1144 | AFI | 1145 +------------------------------------+ 1146 | SAFI | 1147 +------------------------------------+ 1148 | NLRI | 1149 +------------------------------------+ 1151 For the first segment to be processed (the most recently added 1152 segment), the 'Target AS Number' is the AS number of the BGPsec 1153 speaker validating the update message. Note that if a BGPsec 1154 speaker uses multiple AS Numbers (e.g., the BGPsec speaker is a 1155 member of a confederation), the AS number used here MUST be the AS 1156 number announced in the OPEN message for the BGP session over 1157 which the BGPsec update was received. 1159 For each other Signature Segment, the 'Target AS Number' is the AS 1160 number in the Secure_Path segment that corresponds to the 1161 Signature Segment added immediately after the one being processed. 1162 (That is, in the Secure_Path segment that corresponds to the 1163 Signature segment that the validator just finished processing.) 1165 The Secure_Path and Signature Segment are obtained from the 1166 BGPsec_Path attribute. The Address Family Identifier (AFI), 1167 Subsequent Address Family Identifier (SAFI), and Network Layer 1168 Reachability Information (NLRI) fields are obtained from the 1169 MP_REACH_NLRI attribute. Additionally, in the Prefix field of the 1170 NLRI (from MP_REACH_NLRI), all of the trailing bits MUST be set to 1171 zero when constructing this sequence. 1173 o (Step III): Use the signature validation algorithm (for the given 1174 algorithm suite) to verify the signature in the current segment. 1175 That is, invoke the signature validation algorithm on the 1176 following three inputs: the value of the Signature field in the 1177 current segment; the digest value computed in Step II above; and 1178 the public key obtained from the valid RPKI data in Step I above. 1179 If the signature validation algorithm determines that the 1180 signature is invalid, then mark the entire Signature_Block as 'Not 1181 Valid' and proceed to the next Signature_Block. If the signature 1182 validation algorithm determines that the signature is valid, then 1183 continue processing Signature Segments (within the current 1184 Signature_Block). 1186 If all Signature Segments within a Signature_Block pass validation 1187 (i.e., all segments are processed and the Signature_Block has not yet 1188 been marked 'Not Valid'), then the Signature_Block is marked as 1189 'Valid'. 1191 If at least one Signature_Block is marked as 'Valid', then the 1192 validation algorithm terminates and the BGPsec update message is 1193 deemed to be 'Valid'. (That is, if a BGPsec update message contains 1194 two Signature_Blocks then the update message is deemed 'Valid' if the 1195 first Signature_Block is marked 'Valid' OR the second Signature_Block 1196 is marked 'Valid'.) 1198 6. Algorithms and Extensibility 1200 6.1. Algorithm Suite Considerations 1202 Note that there is currently no support for bilateral negotiation 1203 (using BGP capabilities) between BGPsec peers to use a particular 1204 (digest and signature) algorithm suite. This is because the 1205 algorithm suite used by the sender of a BGPsec update message must be 1206 understood not only by the peer to whom it is directly sending the 1207 message, but also by all BGPsec speakers to whom the route 1208 advertisement is eventually propagated. Therefore, selection of an 1209 algorithm suite cannot be a local matter negotiated by BGP peers, but 1210 instead must be coordinated throughout the Internet. 1212 To this end, a mandatory algorithm suites document exists which 1213 specifies a mandatory-to-use 'current' algorithm suite for use by all 1214 BGPsec speakers [I-D.ietf-sidr-bgpsec-algs]. 1216 We anticipate that, in the future, the mandatory algorithm suites 1217 document will be updated to specify a transition from the 'current' 1218 algorithm suite to a 'new' algorithm suite. During the period of 1219 transition (likely a small number of years), all BGPsec update 1220 messages SHOULD simultaneously use both the 'current' algorithm suite 1221 and the 'new' algorithm suite. (Note that Section 3 and Section 4 1222 specify how the BGPsec_Path attribute can contain signatures, in 1223 parallel, for two algorithm suites.) Once the transition is 1224 complete, use of the old 'current' algorithm will be deprecated, use 1225 of the 'new' algorithm will be mandatory, and a subsequent 'even 1226 newer' algorithm suite may be specified as recommended to implement. 1227 Once the transition has successfully been completed in this manner, 1228 BGPsec speakers SHOULD include only a single Signature_Block 1229 (corresponding to the 'new' algorithm). 1231 6.2. Extensibility Considerations 1233 This section discusses potential changes to BGPsec that would require 1234 substantial changes to the processing of the BGPsec_Path and thus 1235 necessitate a new version of BGPsec. Examples of such changes 1236 include: 1238 o A new type of signature algorithm that produces signatures of 1239 variable length 1241 o A new type of signature algorithm for which the number of 1242 signatures in the Signature_Block is not equal to the number of 1243 ASes in the Secure_Path (e.g., aggregate signatures) 1245 o Changes to the data that is protected by the BGPsec signatures 1246 (e.g., attributes other than the AS path) 1248 In the case that such a change to BGPsec were deemed desirable, it is 1249 expected that a subsequent version of BGPsec would be created and 1250 that this version of BGPsec would specify a new BGP path attribute, 1251 let's call it BGPsec_Path_Two, which is designed to accommodate the 1252 desired changes to BGPsec. In such a case, the mandatory algorithm 1253 suites document would be updated to specify algorithm suites 1254 appropriate for the new version of BGPsec. 1256 At this point a transition would begin which is analogous to the 1257 algorithm transition discussed in Section 6.1. During the transition 1258 period all BGPsec speakers should simultaneously include both the 1259 BGPsec_Path attribute and the new BGPsec_Path_Two attribute. Once 1260 the transition is complete, the use of BGPsec_Path could then be 1261 deprecated, at which point BGPsec speakers should include only the 1262 new BGPsec_Path_Two attribute. Such a process could facilitate a 1263 transition to a new BGPsec semantics in a backwards compatible 1264 fashion. 1266 7. Security Considerations 1268 For a discussion of the BGPsec threat model and related security 1269 considerations, please see RFC 7132 [RFC7132]. 1271 7.1. Security Guarantees 1273 When used in conjunction with Origin Validation (see RFC 6483 1274 [RFC6483] and RFC 6811 [RFC6811]), a BGPsec speaker who receives a 1275 valid BGPsec update message, containing a route advertisement for a 1276 given prefix, is provided with the following security guarantees: 1278 o The origin AS number corresponds to an autonomous system that has 1279 been authorized, in the RPKI, by the IP address space holder to 1280 originate route advertisements for the given prefix. 1282 o For each AS in the path, a BGPsec speaker authorized by the holder 1283 of the AS number intentionally chose (in accordance with local 1284 policy) to propagate the route advertisement to the subsequent AS 1285 in the path. 1287 That is, the recipient of a valid BGPsec update message is assured 1288 that the update propagated via the sequence of ASes listed in the 1289 Secure_Path portion of the BGPsec_Path attribute. (It should be 1290 noted that BGPsec does not offer any guarantee that the data packets 1291 would flow along the indicated path; it only guarantees that the BGP 1292 update conveying the path indeed propagated along the indicated 1293 path.) Furthermore, the recipient is assured that this path 1294 terminates in an autonomous system that has been authorized by the IP 1295 address space holder as a legitimate destination for traffic to the 1296 given prefix. 1298 Note that although BGPsec provides a mechanism for an AS to validate 1299 that a received update message has certain security properties, the 1300 use of such a mechanism to influence route selection is completely a 1301 matter of local policy. Therefore, a BGPsec speaker can make no 1302 assumptions about the validity of a route received from an external 1303 BGPsec peer. That is, a compliant BGPsec peer may (depending on the 1304 local policy of the peer) send update messages that fail the validity 1305 test in Section 5. Thus, a BGPsec speaker MUST completely validate 1306 all BGPsec update messages received from external peers. (Validation 1307 of update messages received from internal peers is a matter of local 1308 policy, see Section 5). 1310 7.2. On the Removal of BGPsec Signatures 1312 There may be cases where a BGPsec speaker deems 'Valid' (as per the 1313 validation algorithm in Section 5.2) a BGPsec update message that 1314 contains both a 'Valid' and a 'Not Valid' Signature_Block. That is, 1315 the update message contains two sets of signatures corresponding to 1316 two algorithm suites, and one set of signatures verifies correctly 1317 and the other set of signatures fails to verify. In this case, the 1318 protocol specifies that a BGPsec speaker choosing to propagate the 1319 route advertisement in such an update message SHOULD add its 1320 signature to each of the Signature_Blocks. Thus the BGPsec speaker 1321 creates a signature using both algorithm suites and creates a new 1322 update message that contains both the 'Valid' and the 'Not Valid' set 1323 of signatures (from its own vantage point). 1325 To understand the reason for such a design decision consider the case 1326 where the BGPsec speaker receives an update message with both a set 1327 of algorithm A signatures which are 'Valid' and a set of algorithm B 1328 signatures which are 'Not Valid'. In such a case it is possible 1329 (perhaps even likely, depending on the state of the algorithm 1330 transition) that some of the BGPsec speaker's peers (or other 1331 entities further 'downstream' in the BGP topology) do not support 1332 algorithm A. Therefore, if the BGPsec speaker were to remove the 1333 'Not Valid' set of signatures corresponding to algorithm B, such 1334 entities would treat the message as though it were unsigned. By 1335 including the 'Not Valid' set of signatures when propagating a route 1336 advertisement, the BGPsec speaker ensures that 'downstream' entities 1337 have as much information as possible to make an informed opinion 1338 about the validation status of a BGPsec update. 1340 Note also that during a period of partial BGPsec deployment, a 1341 'downstream' entity might reasonably treat unsigned messages 1342 differently from BGPsec updates that contain a single set of 'Not 1343 Valid' signatures. That is, by removing the set of 'Not Valid' 1344 signatures the BGPsec speaker might actually cause a downstream 1345 entity to 'upgrade' the status of a route advertisement from 'Not 1346 Valid' to unsigned. Finally, note that in the above scenario, the 1347 BGPsec speaker might have deemed algorithm A signatures 'Valid' only 1348 because of some issue with RPKI state local to its AS (for example, 1349 its AS might not yet have obtained a CRL indicating that a key used 1350 to verify an algorithm A signature belongs to a newly revoked 1351 certificate). In such a case, it is highly desirable for a 1352 downstream entity to treat the update as 'Not Valid' (due to the 1353 revocation) and not as 'unsigned' (which would happen if the 'Not 1354 Valid' Signature_Blocks were removed). 1356 A similar argument applies to the case where a BGPsec speaker (for 1357 some reason such as lack of viable alternatives) selects as its best 1358 path (to a given prefix) a route obtained via a 'Not Valid' BGPsec 1359 update message. In such a case, the BGPsec speaker should propagate 1360 a signed BGPsec update message, adding its signature to the 'Not 1361 Valid' signatures that already exist. Again, this is to ensure that 1362 'downstream' entities are able to make an informed decision and not 1363 erroneously treat the route as unsigned. It should also be noted 1364 that due to possible differences in RPKI data observed at different 1365 vantage points in the network, a BGPsec update deemed 'Not Valid' at 1366 an upstream BGPsec speaker may be deemed 'Valid' by another BGP 1367 speaker downstream. 1369 Indeed, when a BGPsec speaker signs an outgoing update message, it is 1370 not attesting to a belief that all signatures prior to its are valid. 1371 Instead it is merely asserting that: 1373 o The BGPsec speaker received the given route advertisement with the 1374 indicated NLRI and Secure_Path; and 1376 o The BGPsec speaker chose to propagate an advertisement for this 1377 route to the peer (implicitly) indicated by the 'Target AS'. 1379 7.3. Mitigation of Denial of Service Attacks 1381 The BGPsec update validation procedure is a potential target for 1382 denial of service attacks against a BGPsec speaker. Here we consider 1383 the mitigation only of denial of service attacks that are specific to 1384 BGPsec. 1386 To mitigate the effectiveness of such denial of service attacks, 1387 BGPsec speakers should implement an update validation algorithm that 1388 performs expensive checks (e.g., signature verification) after 1389 performing less expensive checks (e.g., syntax checks). The 1390 validation algorithm specified in Section 5.2 was chosen so as to 1391 perform checks which are likely to be expensive after checks that are 1392 likely to be inexpensive. However, the relative cost of performing 1393 required validation steps may vary between implementations, and thus 1394 the algorithm specified in Section 5.2 may not provide the best 1395 denial of service protection for all implementations. 1397 Additionally, sending update messages with very long AS paths (and 1398 hence a large number of signatures) is a potential mechanism to 1399 conduct denial of service attacks. For this reason, it is important 1400 that an implementation of the validation algorithm stops attempting 1401 to verify signatures as soon as an invalid signature is found. (This 1402 ensures that long sequences of invalid signatures cannot be used for 1403 denial of service attacks.) Furthermore, implementations can 1404 mitigate such attacks by only performing validation on update 1405 messages that, if valid, would be selected as the best path. That 1406 is, if an update message contains a route that would lose out in best 1407 path selection for other reasons (e.g., a very long AS path) then it 1408 is not necessary to determine the BGPsec-validity status of the 1409 route. 1411 7.4. Additional Security Considerations 1413 The mechanism of setting the pCount field to zero is included in this 1414 specification to enable route servers in the control path to 1415 participate in BGPsec without increasing the effective length of the 1416 AS-PATH. However, entities other than route servers could 1417 conceivably use this mechanism (set the pCount to zero) to attract 1418 traffic (by reducing the effective length of the AS-PATH) 1419 illegitimately. This risk is largely mitigated if every BGPsec 1420 speaker drops incoming update messages that set pCount to zero but 1421 come from a peer that is not a route server. However, note that a 1422 recipient of a BGPsec update message within which an upstream entity 1423 two or more hops away has set pCount to zero is unable to verify for 1424 themselves whether pCount was set to zero legitimately. 1426 BGPsec does not provide protection against attacks at the transport 1427 layer. As with any BGP session, an adversary on the path between a 1428 BGPsec speaker and its peer is able to perform attacks such as 1429 modifying valid BGPsec updates to cause them to fail validation, 1430 injecting (unsigned) BGP update messages without BGPsec_Path 1431 attributes, injecting BGPsec update messages with BGPsec_Path 1432 attributes that fail validation, or causing the peer to tear-down the 1433 BGP session. The use of BGPsec does nothing to increase the power of 1434 an on-path adversary -- in particular, even an on-path adversary 1435 cannot cause a BGPsec speaker to believe a BGPsec-invalid route is 1436 valid. However, as with any BGP session, BGPsec sessions SHOULD be 1437 protected by appropriate transport security mechanisms. 1439 8. IANA Considerations 1441 This document registers a new capability in the registry of BGP 1442 Capabilities. The description for the new capability is "BGPsec 1443 Capability". The reference for the new capability is this document 1444 (i.e., the RFC that replaces draft-ietf-sidr-bgpsec-protocol), see 1445 Section 2.1. 1447 This document registers a new path attribute in the registry of BGP 1448 Path Attributes. The code for this new attribute is "BGPsec_Path". 1449 The reference for the new capability is this document (i.e., the RFC 1450 that replaces draft-ietf-sidr-bgpsec-protocol), see Section 3. 1452 This document does not create any new IANA registries. 1454 9. Contributors 1456 9.1. Authors 1458 Rob Austein 1459 Dragon Research Labs 1460 sra@hactrn.net 1462 Steven Bellovin 1463 Columbia University 1464 smb@cs.columbia.edu 1466 Randy Bush 1467 Internet Initiative Japan 1468 randy@psg.com 1470 Russ Housley 1471 Vigil Security 1472 housley@vigilsec.com 1474 Matt Lepinski 1475 New College of Florida 1476 mlepinski@ncf.edu 1478 Stephen Kent 1479 BBN Technologies 1480 kent@bbn.com 1482 Warren Kumari 1483 Google 1484 warren@kumari.net 1486 Doug Montgomery 1487 USA National Institute of Standards and Technology 1488 dougm@nist.gov 1490 Kotikalapudi Sriram 1491 USA National Institute of Standards and Technology 1492 kotikalapudi.sriram@nist.gov 1494 Samuel Weiler 1495 Parsons 1496 weiler+ietf@watson.org 1498 9.2. Acknowledgements 1500 The authors would like to thank Michael Baer, Luke Berndt, Oliver 1501 Borchert, Wes George, Jeff Haas, Sharon Goldberg, Ed Kern, David 1502 Mandelberg, Doug Maughan, Pradosh Mohapatra, Chris Morrow, Russ 1503 Mundy, Sandy Murphy, Keyur Patel, Mark Reynolds, Heather Schiller, 1504 Jason Schiller, John Scudder, Ruediger Volk and David Ward for their 1505 valuable input and review. 1507 10. References 1509 10.1. Normative References 1511 [I-D.ietf-idr-bgp-extended-messages] 1512 Bush, R., Patel, K., and D. Ward, "Extended Message 1513 support for BGP", draft-ietf-idr-bgp-extended-messages-13 1514 (work in progress), June 2016. 1516 [I-D.ietf-sidr-bgpsec-algs] 1517 Turner, S., "BGPsec Algorithms, Key Formats, & Signature 1518 Formats", draft-ietf-sidr-bgpsec-algs-15 (work in 1519 progress), April 2016. 1521 [I-D.ietf-sidr-bgpsec-pki-profiles] 1522 Reynolds, M., Turner, S., and D. Kent, "A Profile for 1523 BGPsec Router Certificates, Certificate Revocation Lists, 1524 and Certification Requests", draft-ietf-sidr-bgpsec-pki- 1525 profiles-18 (work in progress), July 2016. 1527 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 1528 Requirement Levels", BCP 14, RFC 2119, 1529 DOI 10.17487/RFC2119, March 1997, 1530 . 1532 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 1533 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 1534 DOI 10.17487/RFC4271, January 2006, 1535 . 1537 [RFC4760] Bates, T., Chandra, R., Katz, D., and Y. Rekhter, 1538 "Multiprotocol Extensions for BGP-4", RFC 4760, 1539 DOI 10.17487/RFC4760, January 2007, 1540 . 1542 [RFC5065] Traina, P., McPherson, D., and J. Scudder, "Autonomous 1543 System Confederations for BGP", RFC 5065, 1544 DOI 10.17487/RFC5065, August 2007, 1545 . 1547 [RFC5492] Scudder, J. and R. Chandra, "Capabilities Advertisement 1548 with BGP-4", RFC 5492, DOI 10.17487/RFC5492, February 1549 2009, . 1551 [RFC6482] Lepinski, M., Kent, S., and D. Kong, "A Profile for Route 1552 Origin Authorizations (ROAs)", RFC 6482, 1553 DOI 10.17487/RFC6482, February 2012, 1554 . 1556 [RFC6793] Vohra, Q. and E. Chen, "BGP Support for Four-Octet 1557 Autonomous System (AS) Number Space", RFC 6793, 1558 DOI 10.17487/RFC6793, December 2012, 1559 . 1561 [RFC7606] Chen, E., Ed., Scudder, J., Ed., Mohapatra, P., and K. 1562 Patel, "Revised Error Handling for BGP UPDATE Messages", 1563 RFC 7606, DOI 10.17487/RFC7606, August 2015, 1564 . 1566 10.2. Informative References 1568 [I-D.ietf-sidr-as-migration] 1569 George, W. and S. Murphy, "BGPSec Considerations for AS 1570 Migration", draft-ietf-sidr-as-migration-05 (work in 1571 progress), April 2016. 1573 [I-D.ietf-sidr-bgpsec-ops] 1574 Bush, R., "BGPsec Operational Considerations", draft-ietf- 1575 sidr-bgpsec-ops-10 (work in progress), June 2016. 1577 [I-D.ietf-sidr-rpki-rtr-rfc6810-bis] 1578 Bush, R. and R. Austein, "The Resource Public Key 1579 Infrastructure (RPKI) to Router Protocol", draft-ietf- 1580 sidr-rpki-rtr-rfc6810-bis-07 (work in progress), March 1581 2016. 1583 [I-D.ietf-sidr-rtr-keying] 1584 Bush, R., Turner, S., and K. Patel, "Router Keying for 1585 BGPsec", draft-ietf-sidr-rtr-keying-12 (work in progress), 1586 June 2016. 1588 [RFC6472] Kumari, W. and K. Sriram, "Recommendation for Not Using 1589 AS_SET and AS_CONFED_SET in BGP", BCP 172, RFC 6472, 1590 DOI 10.17487/RFC6472, December 2011, 1591 . 1593 [RFC6480] Lepinski, M. and S. Kent, "An Infrastructure to Support 1594 Secure Internet Routing", RFC 6480, DOI 10.17487/RFC6480, 1595 February 2012, . 1597 [RFC6483] Huston, G. and G. Michaelson, "Validation of Route 1598 Origination Using the Resource Certificate Public Key 1599 Infrastructure (PKI) and Route Origin Authorizations 1600 (ROAs)", RFC 6483, DOI 10.17487/RFC6483, February 2012, 1601 . 1603 [RFC6811] Mohapatra, P., Scudder, J., Ward, D., Bush, R., and R. 1604 Austein, "BGP Prefix Origin Validation", RFC 6811, 1605 DOI 10.17487/RFC6811, January 2013, 1606 . 1608 [RFC7132] Kent, S. and A. Chi, "Threat Model for BGP Path Security", 1609 RFC 7132, DOI 10.17487/RFC7132, February 2014, 1610 . 1612 Authors' Addresses 1614 Matthew Lepinski (editor) 1615 NCF 1616 5800 Bay Shore Road 1617 Sarasota FL 34243 1618 USA 1620 Email: mlepinski@ncf.edu 1622 Kotikalapudi Sriram (editor) 1623 NIST 1624 100 Bureau Drive 1625 Gaithersburg MD 20899 1626 USA 1628 Email: kotikalapudi.sriram@nist.gov