idnits 2.17.1 draft-ietf-idr-flow-spec-02.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** It looks like you're using RFC 3978 boilerplate. You should update this to the boilerplate described in the IETF Trust License Policy document (see https://trustee.ietf.org/license-info), which is required now. -- Found old boilerplate from RFC 3978, Section 5.1 on line 21. -- Found old boilerplate from RFC 3978, Section 5.5, updated by RFC 4748 on line 826. -- Found old boilerplate from RFC 3979, Section 5, paragraph 1 on line 837. -- Found old boilerplate from RFC 3979, Section 5, paragraph 2 on line 844. -- Found old boilerplate from RFC 3979, Section 5, paragraph 3 on line 850. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- == No 'Intended status' indicated for this document; assuming Proposed Standard Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust Copyright Line does not match the current year -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (September 3, 2008) is 5714 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) -- Looks like a reference, but probably isn't: '137' on line 403 -- Looks like a reference, but probably isn't: '139' on line 403 ** Obsolete normative reference: RFC 793 (Obsoleted by RFC 9293) ** Obsolete normative reference: RFC 5226 (Obsoleted by RFC 8126) Summary: 3 errors (**), 0 flaws (~~), 2 warnings (==), 9 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 IDR Working Group P. Marques 3 Internet-Draft N. Sheth 4 Expires: March 7, 2009 R. Raszuk 5 B. Greene 6 Juniper Networks 7 J. Mauch 8 NTT/Verio 9 D. McPherson 10 Arbor Networks 11 September 3, 2008 13 Dissemination of flow specification rules 14 draft-ietf-idr-flow-spec-02 16 Status of this Memo 18 By submitting this Internet-Draft, each author represents that any 19 applicable patent or other IPR claims of which he or she is aware 20 have been or will be disclosed, and any of which he or she becomes 21 aware will be disclosed, in accordance with Section 6 of BCP 79. 23 Internet-Drafts are working documents of the Internet Engineering 24 Task Force (IETF), its areas, and its working groups. Note that 25 other groups may also distribute working documents as Internet- 26 Drafts. 28 Internet-Drafts are draft documents valid for a maximum of six months 29 and may be updated, replaced, or obsoleted by other documents at any 30 time. It is inappropriate to use Internet-Drafts as reference 31 material or to cite them other than as "work in progress." 33 The list of current Internet-Drafts can be accessed at 34 http://www.ietf.org/ietf/1id-abstracts.txt. 36 The list of Internet-Draft Shadow Directories can be accessed at 37 http://www.ietf.org/shadow.html. 39 This Internet-Draft will expire on March 7, 2009. 41 Abstract 43 This document defines a new BGP NLRI encoding format that can be used 44 to distribute traffic flow specifications. This allows the routing 45 system to propagate information regarding more-specific components of 46 the traffic aggregate defined by an IP destination prefix. 48 Additionally it defines two applications of that encoding format. 49 One that can be used to automate inter-domain coordination of traffic 50 filtering, such as what is required in order to mitigate 51 (distributed) denial of service attacks. And a second application to 52 traffic filtering in the context of a BGP/MPLS VPN service. 54 The information is carried via the Border Gateway Protocol (BGP), 55 thereby reusing protocol algorithms, operational experience and 56 administrative processes such as inter-provider peering agreements. 58 Table of Contents 60 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 4 61 2. Flow specifications . . . . . . . . . . . . . . . . . . . . . 6 62 3. Dissemination of Information . . . . . . . . . . . . . . . . . 7 63 4. Traffic filtering . . . . . . . . . . . . . . . . . . . . . . 13 64 4.1. Order of traffic filtering rules . . . . . . . . . . . . . 14 65 5. Validation procedure . . . . . . . . . . . . . . . . . . . . . 15 66 6. Traffic Filtering Actions . . . . . . . . . . . . . . . . . . 17 67 7. Traffic filtering in RFC2547bis networks . . . . . . . . . . . 19 68 8. Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . 20 69 9. Security considerations . . . . . . . . . . . . . . . . . . . 21 70 10. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 22 71 11. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 23 72 12. Normative References . . . . . . . . . . . . . . . . . . . . . 24 73 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 25 74 Intellectual Property and Copyright Statements . . . . . . . . . . 27 76 1. Introduction 78 Modern IP routers contain both the capability to forward traffic 79 according to aggregate IP prefixes as well as to classify, shape, 80 limit filter or redirect packets based on administratively defined 81 policies. 83 While forwarding information is, typically, dynamically signaled 84 across the network via routing protocols, there is no agreed upon 85 mechanism to dynamically signal flows across autonomous-systems. 87 For several applications, it may be necessary to exchange control 88 information pertaining to aggregated traffic flow definitions which 89 cannot be expressed using destination address prefixes only. 91 An aggregated traffic flow is considered to be an n-tuple consisting 92 of several matching criteria such as source and destination address 93 prefixes, IP protocol and transport protocol port numbers. 95 The intention of this document is to define a general procedure to 96 encode such flow specification rules as a BGP [RFC4271] NLRI which 97 can be reused for several different control applications. 98 Additionally, we define the required mechanisms to utilize this 99 definition to the problem of immediate concern to the authors: intra 100 and inter provider distribution of traffic filtering rules to filter 101 (Distributed) Denial of Service (DoS) attacks. 103 By expanding routing information with flow specifications, the 104 routing system can take advantage of the ACL/firewall capabilities in 105 the router's forwarding path. Flow specifications can be seen as 106 more specific routing entries to an unicast prefix and are expected 107 to depend upon the existing unicast data information. 109 A flow specification received from a external autonomous-system will 110 need to be validated against unicast routing before being accepted. 111 If the aggregate traffic flow defined by the unicast destination 112 prefix is forwarded to a given BGP peer, then the local system can 113 safely install more specific flow rules which result in different 114 forwarding behavior, as requested by this system. 116 The choice of BGP as the carrier of this control information is also 117 justifiable by the fact that the key issues in terms of complexity 118 are problems which are common to unicast route distribution and have 119 already been solved in the current environment. 121 From an algorithmic perspective, the main problem that presents 122 itself is the loop-free distribution of pairs from 123 one originator to N ingresses. The key, in this particular instance, 124 being a flow specification. 126 From an operational perspective, the utilization of BGP as the 127 carrier for this information, allows a network service provider to 128 reuse both internal route distribution infrastructure (e.g.: route 129 reflector or confederation design) and existing external 130 relationships (e.g.: inter-domain BGP sessions to a customer 131 network). 133 While it is certainly possible to address this problem using other 134 mechanisms, the authors believe that this solution offers the 135 substantial advantage of being an incremental addition to deployed 136 mechanisms. 138 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 139 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 140 document are to be interpreted as described in RFC 2119 [RFC2119]. 142 2. Flow specifications 144 A flow specification is an n-tuple consisting on several matching 145 criteria that can be applied to IP traffic. A given IP packet is 146 said to match the defined flow if it matches all the specified 147 criteria. 149 A given flow may be associated with a set of attributes, depending on 150 the particular application, such attributes may or may not include 151 reachability information (i.e. NEXT_HOP). Well-known or AS-specific 152 community attributes can be used to encode a set of predeterminate 153 actions. 155 A particular application is identified by a specific (AFI, SAFI) pair 156 [RFC4760] and corresponds to a distinct set of RIBs. Those RIBs 157 should be treated independently from each other in order to assure 158 non-interference between distinct applications. 160 BGP itself treats the NLRI as an opaque key to an entry in its 161 databases. Entries that are placed in the Loc-RIB are then 162 associated with a given set of semantics which is application 163 dependent. This is consistent with existing BGP applications. For 164 instance IP unicast routing (AFI=1, SAFI=1) and IP multicast reverse- 165 path information (AFI=1, SAFI=2) are handled by BGP without any 166 particular semantics being associated with them until installed in 167 the Loc-RIB. 169 Standard BGP policy mechanisms, such as UPDATE filtering by NLRI 170 prefix and community matching, SHOULD apply to the newly defined 171 NLRI-type. Network operators can also control propagation of such 172 routing updates by enabling or disabling the exchange of a particular 173 (AFI, SAFI) pair on a given BGP peering session. 175 3. Dissemination of Information 177 We define a "Flow Specification" NLRI type that may include several 178 components such as destination prefix, source prefix, protocol, 179 ports, etc. This NLRI is treated as an opaque bit string prefix by 180 BGP. Each bit string identifies a key to a database entry which a 181 set of attributes can be associated with. 183 This NLRI information is encoded using MP_REACH_NLRI and 184 MP_UNREACH_NLRI attributes as defined in RFC4760 [RFC4760]. Whenever 185 the corresponding application does not require Next Hop information, 186 this shall be encoded as a 0 octet length Next Hop in the 187 MP_REACH_NLRI attribute and ignored on receipt. 189 The NLRI field of the MP_REACH_NLRI and MP_UNREACH_NLRI is encoded as 190 a 1 or 2 octet NLRI length field followed by a variable length NLRI 191 value. The NLRI length is expressed in octets. 193 +------------------------------+ 194 | length (0xnn or 0xfn nn) | 195 +------------------------------+ 196 | NLRI value (variable) | 197 +------------------------------+ 199 flow-spec NLRI 201 If the NLRI length value is smaller than 240 (0xf0 hex), the length 202 field can be encoded as a single octet. Otherwise, it is encoded as 203 a extended length 2 octet value in which the most significant nibble 204 of the first byte is all ones. 206 The Flow Specification NLRI-type consists of several optional 207 subcomponents. A specific packet is considered to match the flow 208 specification when it matches the intersection (AND) of all the 209 components present in the specification. 211 The following component types are defined: 213 Type 1 - Destination Prefix 215 Encoding: 217 Defines the destination prefix to match. Prefixes are encoded 218 as in BGP UPDATE messages, a length in bits is followed by 219 enough octets to contain the prefix information. 221 Type 2 - Source Prefix 223 Encoding: 225 Defines the source prefix to match. 227 Type 3 - IP Protocol 229 Encoding: 231 Contains a set of {operator, value} pairs that are used to 232 match IP protocol value byte in IP packets. 234 The operator byte is encoded as: 236 7 6 5 4 3 2 1 0 237 +---+---+---+---+---+---+---+---+ 238 | e | a | len | 0 |lt |gt |eq | 239 +---+---+---+---+---+---+---+---+ 241 Numeric operator 243 + End of List bit. Set in the last {op, value} pair in the 244 list. 246 + And bit. If unset the previous term is logically ORed with 247 the current one. If set the operation is a logical AND. It 248 should be unset in the first operator byte of a sequence. 249 The AND operator has higher priority than OR for the 250 purposes of evaluating logical expressions. 252 + The length of value field for this operand is given as (1 << 253 len). 255 + Lt - less than comparison between data and value. 257 + gt - greater than comparison between data and value. 259 + eq - equality between data and value. 261 The bits lt, gt, and eq can be combined to produce "less or 262 equal", "greater or equal" and inequality values. 264 Type 4 - Port 266 Encoding: 267 Defines a list of {operation, value} pairs that matches source 268 OR destination TCP/UDP ports. This list is encoded using the 269 numeric operand format defined above. Values are encoded as 1 270 or 2 byte quantities. 272 Type 5 - Destination port 274 Encoding: 276 Defines a list of {operation, value} pairs used to match the 277 destination port of a TCP or UDP packet. Values are encoded as 278 1 or 2 byte quantities. 280 Type 6 - Source port 282 Encoding: 284 Defines a list of {operation, value} pairs used to match the 285 source port of a TCP or UDP packet. Values are encoded as 1 or 286 2 byte quantities. 288 Type 7 - ICMP type 290 Encoding: 292 Defines a list of {operation, value} pairs used to match the 293 type field of an icmp packet. Values are encoded using a 294 single byte. 296 Type 8 - ICMP code 298 Encoding: 300 Defines a list of {operation, value} pairs used to match the 301 code field of an icmp packet. Values are encoded using a 302 single byte. 304 Type 9 - TCP flags 306 Encoding: 308 Bitmask values are encoded using a single byte, using the bit 309 definitions specified in the TCP header format [RFC0793]. 311 This type uses the bitmask operand format, which differs from 312 the numeric operator format in the lower nibble. 314 7 6 5 4 3 2 1 0 315 +---+---+---+---+---+---+---+---+ 316 | e | a | len | 0 | 0 |not| m | 317 +---+---+---+---+---+---+---+---+ 319 + Top nibble: (End of List bit, And bit and Length field), as 320 defined for in the numeric operator format. 322 + Not bit. If set, logical negation of operation. 324 + Match bit. If set this is a bitwise match operation defined 325 as "(data & value) == value"; if unset (data & value) 326 evaluates to true if and of the bits in the value mask are 327 set in the data. 329 Type 10 - Packet length 331 Encoding: 333 Match on the total IP packet length (excluding L2 but including 334 IP header). Values are encoded using as 1 or 2 byte 335 quantities. 337 Type 11 - DSCP 339 Encoding: 341 Defines a list of {operation, value} pairs used to match the IP 342 TOS octet. 344 Type 12 - Fragment 346 Encoding: 348 Uses bitmask operand format defined above. 350 Bitmask values: 352 + Bit 0 - Dont fragment 354 + Bit 1 - Is a fragment 356 + Bit 2 - First fragment 358 + Bit 3 - Last fragment 360 Flow specification components must follow strict type ordering. A 361 given component type may or may not be present in the specification, 362 but if present it MUST precede any component of higher numeric type 363 value. 365 If a given component type within a prefix in unknown, the prefix in 366 question cannot be used for traffic filtering purposes by the 367 receiver. Since a Flow Specification as the semantics of a logical 368 AND of all components, if a component is FALSE by definition it 369 cannot be applied. However for the purposes of BGP route propagation 370 this prefix should still be transmitted since BGP route distribution 371 is independent on NLRI semantics. 373 Flow specification components are to be interpreted as a bit match at 374 a given packet offset. When more than one component in a flow 375 specification tests the same packet offset the behavior is 376 undetermined. 378 The encoding is chosen in order to account for future 379 extensibility. 381 An example of a Flow Specification encoding for: "all packets to 382 10.0.1/24 and TCP port 25". 384 +------------------+----------+----------+ 385 | destination | proto | port | 386 +------------------+----------+----------+ 387 | 0x01 18 0a 00 01 | 03 81 06 | 04 81 19 | 388 +------------------+----------+----------+ 390 Decode for protocol: 392 +-------+----------+------------------------------+ 393 | Value | | | 394 +-------+----------+------------------------------+ 395 | 0x03 | type | | 396 | | | | 397 | 0x81 | operator | end-of-list, value size=1, = | 398 | | | | 399 | 0x06 | value | | 400 +-------+----------+------------------------------+ 402 An example of a Flow Specification encoding for: "all packets to 403 10.0.1/24 from 192/8 and port {range [137, 139] or 8080}". 405 +------------------+----------+-------------------------+ 406 | destination | source | port | 407 +------------------+----------+-------------------------+ 408 | 0x01 18 0a 01 01 | 02 08 c0 | 04 03 89 45 8b 91 1f 90 | 409 +------------------+----------+-------------------------+ 410 Decode for port: 412 +--------+----------+------------------------------+ 413 | Value | | | 414 +--------+----------+------------------------------+ 415 | 0x04 | type | | 416 | | | | 417 | 0x03 | operator | size=1, >= | 418 | | | | 419 | 0x89 | value | 137 | 420 | | | | 421 | 0x45 | operator | &, value size=1, <= | 422 | | | | 423 | 0x8b | value | 139 | 424 | | | | 425 | 0x91 | operator | end-of-list, value-size=2, = | 426 | | | | 427 | 0x1f90 | value | 8080 | 428 +--------+----------+------------------------------+ 430 This constitutes a NLRI with an NLRI length of 16 octets. 432 Implementations wishing to exchange flow specification rules MUST use 433 BGP's Capability Advertisement facility to exchange the Multiprotocol 434 Extension Capability Code (Code 1) as defined in RFC4760 [RFC4760]. 435 The (AFI, SAFI) pair carried in the Multiprotocol Extension 436 capability MUST be the same as the one used to identify a particular 437 application that uses this NLRI-type. 439 4. Traffic filtering 441 Traffic filtering policies have been traditionally considered to be 442 relatively static. 444 The popularity of traffic-based denial of service (DoS) attacks, 445 which often requires the network operator to be able to use traffic 446 filters for detection and mitigation, brings with it requirements 447 that are not fully satisfied by existing tools. 449 Increasingly, DoS mitigation, requires coordination among several 450 Service Providers, in order to be able to identify traffic source(s) 451 and because the volumes of traffic may be such that they will 452 otherwise significantly affect the performance of the network. 454 Several techniques are currently used to control traffic filtering of 455 DoS attacks. Among those, one of the most common is to inject 456 unicast route advertisements corresponding to a destination prefix 457 being attacked. One variant of this technique marks such route 458 advertisements with a community that gets translated into a discard 459 next-hop by the receiving router. Other variants, attract traffic to 460 a particular node that serves as a deterministic drop point. 462 Using unicast routing advertisements to distribute traffic filtering 463 information has the advantage of using the existing infrastructure 464 and inter-as communication channels. This can allow, for instance, 465 for a service provider to accept filtering requests from customers 466 for address space they own. 468 There are several drawbacks, however. An issue that is immediately 469 apparent is the granularity of filtering control: only destination 470 prefixes may be specified. Another area of concern is the fact that 471 filtering information is intermingled with routing information. 473 The mechanism defined in this document is designed to address these 474 limitations. We use the flow specification NLRI defined above to 475 convey information about traffic filtering rules for traffic that 476 should be discarded. 478 This mechanism is designed to, primarily, allow an upstream 479 autonomous system to perform inbound filtering, in their ingress 480 routers of traffic that a given downstream AS wishes to drop. 482 In order to achieve that goal, we define an application specific NLRI 483 identifier (AFI=1, SAFI=133) along with specific semantic rules. 485 BGP routing updates containing this identifier use the flow 486 specification NLRI encoding to convey particular aggregated flows 487 that require special treatment. 489 Flow routing information received via this (afi, safi) pair is 490 subject to the validation procedure detailed bellow. 492 4.1. Order of traffic filtering rules 494 With traffic filtering rules, more than one rule may match a 495 particular traffic flow. Thus it is necessary to define the order at 496 which rules get matched and applied to a particular traffic flow. 497 This ordering function must be such that it must not depend on the 498 arrival order of the flow specifications rules and must be constant 499 in the network. 501 We choose to order traffic filtering rules such that the order of two 502 flow specifications is given by the comparison of NLRI key byte 503 strings as defined by the memcmp() function is the ISO C standard. 505 Given the way that flow specifications are encoded this results in a 506 flow with a less-specific destination IP prefix being considered 507 less-than (and thus match before) a flow specification with a more- 508 specific destination IP prefix. 510 This matches an application model where the user may want to define a 511 restriction that affects an aggregate of traffic and a subsequent 512 rule that applies only to a subset of that. 514 A flow-specification without a destination IP prefix is considered to 515 match after all flow-specifications that contain an IP destination 516 prefix. 518 5. Validation procedure 520 Flow specifications received from a BGP peer and which are accepted 521 in the respective Adj-RIB-In are used as input to the route selection 522 process. Although the forwarding attributes of two routes for the 523 same Flow Specification prefix may be the same, BGP is still required 524 to perform its path selection algorithm in order to select the 525 correct set of attributes to advertise. 527 The first step of the BGP Route Selection procedure (section 9.1.2) 528 is to exclude from the selection procedure routes that are considered 529 non-feasible. In the context of IP routing information this step is 530 used to validate that the NEXT_HOP attribute of a given route is 531 resolvable. 533 The concept can be extended, in the case of Flow Specification NLRI, 534 to allow other validation procedures. 536 A flow specification NLRI must be validated such that it is 537 considered feasible if and only if: 539 a) The originator of the flow specification matches the originator of 540 the best-match unicast route for the destination prefix embedded 541 in the flow specification. 543 b) There are no more-specific unicast routes, when compared with the 544 flow destination prefix, that have been received from a different 545 neighboring AS than the best-match unicast route, which has been 546 determined in step a). 548 By originator of a BGP route, we mean either the BGP originator path 549 attribute, as used by route reflection, or the transport address of 550 the BGP peer, if this path attribute is not present. 552 The underlying concept is that the neighboring AS that advertises the 553 best unicast route for a destination is allowed to advertise flow- 554 spec information that conveys a more or equally specific destination 555 prefix. This, as long as there are no more-specific unicast routes, 556 received from a different neighbor AS, which would be affected by 557 that filtering rule. 559 The neighboring AS is the immediate destination of the traffic 560 described by the Flow Specification. If it requests these flows to 561 be dropped that request can be honored without concern that it 562 represents a denial of service in itself. Supposedly, the traffic is 563 being dropped by the downstream autonomous-system and there is no 564 added value in carrying the traffic to it. 566 BGP implementations MUST also enforce that the AS_PATH attribute of a 567 route received via eBGP contains the neighboring AS in the left-most 568 position of the AS_PATH attribute. While this rule is optional in 569 the BGP specification, it becomes necessary to enforce it for 570 security reasons. 572 6. Traffic Filtering Actions 574 This specification defines a minimum set of filtering actions that it 575 standardizes as BGP extended community values [RFC4360]. This is not 576 ment to be an inclusive list of all the possible actions but only a 577 subset that can be interpreted consistently across the network. 579 Implementations should provide mechanisms that map an arbitrary bgp 580 community value (normal or extended) to filtering actions that 581 require different mappings in different systems in the network. For 582 instance, providing packets with a worse than best-effort per-hop 583 behavior is a functionality that is likely to be implemented 584 differently in different systems and for which no standard behavior 585 is currently known. Rather than attempting to define it here, this 586 can be accomplished by mapping a user defined community value to 587 platform / network specific behavior via user configuration. 589 The default action for a traffic filtering flow specification is to 590 accept IP traffic that matches that particular rule. 592 The following extended community values can be used to specify 593 particular actions. 595 +--------+--------------------+--------------------------+ 596 | type | extended community | encoding | 597 +--------+--------------------+--------------------------+ 598 | 0x8006 | traffic-rate | 2-byte as#, 4-byte float | 599 | | | | 600 | 0x8007 | traffic-action | bitmask | 601 | | | | 602 | 0x8008 | redirect | 6-byte Route Target | 603 +--------+--------------------+--------------------------+ 605 Traffic-rate The traffic-rate extended community uses the same 606 encoding as the "Link Bandwidth" [RFC4360] extended community. 607 The rate is is expressed as 4 octets in IEEE floating point 608 format, units being bytes per second. A traffic-rate of 0 should 609 result on all traffic for the particular flow to be discarded. 611 Traffic-action The traffic-action extended community consists of 6 612 bytes of which only the 2 least significant bits of the 6th byte 613 (from left to right) are currently defined. 615 * Terminal action (bit 0). When this bit is set the traffic 616 filtering engine will apply any subsequent filtering rules (as 617 defined by the ordering procedure). If not set the evaluation 618 of the traffic filter stops when this rule is applied. 620 * Sample (bit 1). Enables traffic sampling and logging for this 621 flow specification. 623 Redirect The redirect extended community allows the traffic to be 624 redirected to a VRF routing instance that list the specified 625 route-target in its import policy. If several local instances 626 match this criteria, the choice between them is a local matter 627 (for example, the instance with the lowest Route Distinguisher 628 value can be elected). The traffic marking extended community 629 instruct a system to modify the DSCP bits of a transiting IP 630 packet to the corresponding value. This extended community is 631 encoded as a sequence of 5 zero bytes followed by the DSCP value. 633 7. Traffic filtering in RFC2547bis networks 635 Provider-based layer 3 VPN networks, such as the ones using an BGP/ 636 MPLS IP VPN [RFC4364] control plane, have different traffic filtering 637 requirements than internet service providers. 639 In these environments, the VPN customer network often has traffic 640 filtering capabilities towards their external network connections 641 (e.g. firewall facing public network connection). Less common is the 642 presence of traffic filtering capabilities between different VPN 643 attachment sites. In an any-to-any connectivity model, which is the 644 default, this means that site to site traffic is unfiltered. 646 In circumstances where a security threat does get propagated inside 647 the VPN customer network, there may not be readily available 648 mechanisms to provide mitigation via traffic filter. 650 This document proposes an additional BGP NLRI type (afi=1, safi=134) 651 value, which can be used to propagate traffic filtering information 652 in a BGP/MPLS VPN environment. 654 The NLRI format for this address family consists of a fixed length 655 Route Distinguisher field (8 bytes) followed by a flow specification, 656 following the encoded defined in this document. The NLRI length 657 field shall includes the both 8 bytes of the Route Distinguisher as 658 well as the subsequent flow specification. 660 Propagation of this NLRI is controlled by matching Route Target 661 extended communities associated with the BGP path advertisement with 662 the VRF import policy, using the same mechanism as described in "BGP/ 663 MPLS IP VPNs" [RFC4364] . 665 Flow specification rules received via this NLRI apply only to traffic 666 that belongs to the VRF(s) in which it is imported. By default, 667 traffic received from a remote PE is switched via an mpls forwarding 668 decision and is not subject to filtering. 670 Contrary to the behavior specified for the non-VPN NLRI, flow rules 671 are accepted by default, when received from remote PE routers. 673 8. Monitoring 675 Traffic filtering applications require monitoring and traffic 676 statistics facilities. While this is an implementation specific 677 choice, implementations SHOULD provide: 679 o A mechanism to log the packet header of filtered traffic, 681 o A mechanism to count the number of matches for a given Flow 682 Specification rule. 684 9. Security considerations 686 Inter-provider routing is based on a web of trust. Neighboring 687 autonomous-systems are trusted to advertise valid reachability 688 information. If this trust model is violated, a neighboring 689 autonomous system may cause a denial of service attack by advertising 690 reachability information for a given prefix for which it does not 691 provide service. 693 As long as traffic filtering rules are restricted to match the 694 corresponding unicast routing paths for the relevant prefixes, the 695 security characteristics of this proposal are equivalent to the 696 existing security properties of BGP unicast routing. 698 Where it not the case, this would open the door to further denial of 699 service attacks. 701 10. IANA Considerations 703 A flow specification consists of a sequence of flow components, which 704 are identified by a an 8-bit component type. Types must be assigned 705 and interpreted uniquely. The current specification defines types 1 706 though 12, with the value 0 being reserved. 708 In order to manage the limited number space and accommodate several 709 usages the following policies defined by RFC 5226 [RFC5226] are used: 711 +--------------+-------------------------------+ 712 | Range | Policy | 713 +--------------+-------------------------------+ 714 | 0 | Invalid value | 715 | | | 716 | [1 .. 12] | Defined by this specification | 717 | | | 718 | [13 .. 127] | Specification Required | 719 | | | 720 | [128 .. 255] | Private Use | 721 +--------------+-------------------------------+ 723 The specification of a particular "flow component type" must clearly 724 identify what is the criteria used to match packets forwarded by the 725 router. This criteria should be meaningful across router hops and 726 not depend on values that change hop-by-hop such as ttl or layer-2 727 encapsulation. 729 The "Traffic-action" extended community defined in this document has 730 6 unused bits which can be used to convey additional meaning. These 731 values should be assigned via IETF Consensus rules only. 733 11. Acknowledgments 735 The authors would like to thank Yakov Rekhter, Dennis Ferguson and 736 Chris Morrow for their comments. 738 Chaitanya Kodeboyina helped design the flow validation procedure. 740 Steven Lin and Jim Washburn ironed out all the details necessary to 741 produce a working implementation. 743 12. Normative References 745 [RFC0793] Postel, J., "Transmission Control Protocol", STD 7, 746 RFC 793, September 1981. 748 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 749 Requirement Levels", BCP 14, RFC 2119, March 1997. 751 [RFC4271] Rekhter, Y., Li, T., and S. Hares, "A Border Gateway 752 Protocol 4 (BGP-4)", RFC 4271, January 2006. 754 [RFC4760] Bates, T., Chandra, R., Katz, D., and Y. Rekhter, 755 "Multiprotocol Extensions for BGP-4", RFC 4760, 756 January 2007. 758 [RFC4360] Sangli, S., Tappan, D., and Y. Rekhter, "BGP Extended 759 Communities Attribute", RFC 4360, February 2006. 761 [RFC4364] Rosen, E. and Y. Rekhter, "BGP/MPLS IP Virtual Private 762 Networks (VPNs)", RFC 4364, February 2006. 764 [RFC5226] Narten, T. and H. Alvestrand, "Guidelines for Writing an 765 IANA Considerations Section in RFCs", BCP 26, RFC 5226, 766 May 2008. 768 Authors' Addresses 770 Pedro Marques 771 Juniper Networks 772 1194 N. Mathilda Ave. 773 Sunnyvale, CA 94089 774 US 776 Email: roque@juniper.net 778 Nischal Sheth 779 Juniper Networks 780 1194 N. Mathilda Ave. 781 Sunnyvale, CA 94089 782 US 784 Email: nsheth@juniper.net 786 Robert Raszuk 787 Juniper Networks 788 1194 N. Mathilda Ave. 789 Sunnyvale, CA 94089 790 US 792 Email: raszuk@juniper.net 794 Barry Greene 795 Juniper Networks 796 1194 N. Mathilda Ave. 797 Sunnyvale, CA 94089 798 US 800 Email: bgreene@juniper.net 802 Jared Mauch 803 NTT/Verio 804 8285 Reese Lane 805 Ann Arbor, MI 48103-9753 806 US 807 Danny McPherson 808 Arbor Networks 810 Email: danny@arbor.net 812 Full Copyright Statement 814 Copyright (C) The IETF Trust (2008). 816 This document is subject to the rights, licenses and restrictions 817 contained in BCP 78, and except as set forth therein, the authors 818 retain all their rights. 820 This document and the information contained herein are provided on an 821 "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS 822 OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND 823 THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS 824 OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF 825 THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED 826 WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. 828 Intellectual Property 830 The IETF takes no position regarding the validity or scope of any 831 Intellectual Property Rights or other rights that might be claimed to 832 pertain to the implementation or use of the technology described in 833 this document or the extent to which any license under such rights 834 might or might not be available; nor does it represent that it has 835 made any independent effort to identify any such rights. Information 836 on the procedures with respect to rights in RFC documents can be 837 found in BCP 78 and BCP 79. 839 Copies of IPR disclosures made to the IETF Secretariat and any 840 assurances of licenses to be made available, or the result of an 841 attempt made to obtain a general license or permission for the use of 842 such proprietary rights by implementers or users of this 843 specification can be obtained from the IETF on-line IPR repository at 844 http://www.ietf.org/ipr. 846 The IETF invites any interested party to bring to its attention any 847 copyrights, patents or patent applications, or other proprietary 848 rights that may cover technology that may be required to implement 849 this standard. Please address the information to the IETF at 850 ietf-ipr@ietf.org.