idnits 2.17.1 draft-raszuk-idr-rfc5575bis-00.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 : ---------------------------------------------------------------------------- ** There is 1 instance of too long lines in the document, the longest one being 25 characters in excess of 72. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == The document seems to lack the recommended RFC 2119 boilerplate, even if it appears to use RFC 2119 keywords -- however, there's a paragraph with a matching beginning. Boilerplate error? (The document does seem to have the reference to RFC 2119 which the ID-Checklist requires). -- The document date (June 24, 2016) is 2863 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 729 -- Looks like a reference, but probably isn't: '139' on line 729 == Missing Reference: 'IEEE.754.1985' is mentioned on line 933, but not defined == Unused Reference: 'RFC4761' is defined on line 1341, but no explicit reference was found in the text == Unused Reference: 'RFC4762' is defined on line 1346, but no explicit reference was found in the text == Unused Reference: 'RFC6241' is defined on line 1366, but no explicit reference was found in the text == Unused Reference: 'RFC6482' is defined on line 1371, but no explicit reference was found in the text == Unused Reference: 'RFC7223' is defined on line 1380, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-idr-bgp-flowspec-oid' is defined on line 1401, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-idr-flowspec-interfaceset' is defined on line 1413, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-idr-flowspec-packet-rate' is defined on line 1435, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-idr-wide-bgp-communities' is defined on line 1440, but no explicit reference was found in the text == Unused Reference: 'RFC6074' is defined on line 1450, but no explicit reference was found in the text == Unused Reference: 'RFC6483' is defined on line 1456, but no explicit reference was found in the text ** Obsolete normative reference: RFC 793 (Obsoleted by RFC 9293) ** Obsolete normative reference: RFC 5226 (Obsoleted by RFC 8126) ** Obsolete normative reference: RFC 5575 (Obsoleted by RFC 8955) ** Obsolete normative reference: RFC 7223 (Obsoleted by RFC 8343) ** Obsolete normative reference: RFC 7674 (Obsoleted by RFC 8955) == Outdated reference: A later version (-02) exists of draft-ietf-idr-bgp-flowspec-label-00 == Outdated reference: A later version (-15) exists of draft-ietf-idr-bgp-flowspec-oid-03 == Outdated reference: A later version (-22) exists of draft-ietf-idr-flow-spec-v6-07 == Outdated reference: A later version (-05) exists of draft-ietf-idr-flowspec-interfaceset-01 == Outdated reference: A later version (-23) exists of draft-ietf-idr-flowspec-l2vpn-04 == Outdated reference: A later version (-02) exists of draft-ietf-idr-flowspec-mpls-match-00 == Outdated reference: A later version (-19) exists of draft-ietf-idr-flowspec-nvo3-00 == Outdated reference: A later version (-01) exists of draft-ietf-idr-flowspec-packet-rate-00 == Outdated reference: A later version (-11) exists of draft-ietf-idr-wide-bgp-communities-02 Summary: 6 errors (**), 0 flaws (~~), 23 warnings (==), 3 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 IDR Working Group R. Raszuk 3 Internet-Draft Bloomberg LP 4 Intended status: Standards Track J. McPherson 5 Expires: December 26, 2016 Verisign 6 J. Mauch 7 NTT America 8 B. Greene 9 Juniper Networks 10 S. Hares 11 Huawei 12 June 24, 2016 14 Dissemination of Flow Specification Rules 15 draft-raszuk-idr-rfc5575bis-00.txt 17 Abstract 19 This document defines a Border Gateway Protocol Network Layer 20 Reachability Information (BGP NLRI) encoding format that can be used 21 to distribute traffic flow specifications. This allows the routing 22 system to propagate information regarding more specific components of 23 the traffic aggregate defined by an IP destination prefix (IPv4, 24 IPv6), MPLS addresses, L2VPN addresses, and NV03 encapsulation of IP 25 addresses. The information is carried via the BGP, thereby reusing 26 protocol algorithms, operational experience, and administrative 27 processes such as inter-provider peering agreements. 29 This document provides the definition of a BGP NLRI which carries 30 traffic flow specification filters, and Extended Community values 31 which encode the actions a routing system can take if a packet 32 matches the traffic flow filters. The specification requires that 33 the BGP Flow Specification traffic filters follows a string ordering, 34 and that the BGP Flow Specification Extended Communities actions are 35 processed in a defined order. This BGP Flow Specification is denoted 36 as BGP Flow Specification version 1. 38 There are three applications of that encoding format: 1) automation 39 of inter-domain coordination of traffic filtering, such as what is 40 required in order to mitigate (distributed) denial-of-service 41 attacks; 2) enable traffic filtering in the context of a BGP/MPLS VPN 42 service, and 3) aid centralized control of traffic in a SDN or NFV 43 context. Some of deployments of these three applications can be 44 handled by the strict ordering of the BGP NLRI traffic flow filters, 45 and the strict actions encoded in the Extended Community Flow 46 Specification actions. Other deployments (especially SDN/NFV) need 47 to be able to allow the user to order the flow specification. 48 Another BGP Flow Specification (version 2) is being defined for user- 49 ordered filters, and user-ordered actions encoded in Wide 50 Communities. 52 Status of This Memo 54 This Internet-Draft is submitted in full conformance with the 55 provisions of BCP 78 and BCP 79. 57 Internet-Drafts are working documents of the Internet Engineering 58 Task Force (IETF). Note that other groups may also distribute 59 working documents as Internet-Drafts. The list of current Internet- 60 Drafts is at http://datatracker.ietf.org/drafts/current/. 62 Internet-Drafts are draft documents valid for a maximum of six months 63 and may be updated, replaced, or obsoleted by other documents at any 64 time. It is inappropriate to use Internet-Drafts as reference 65 material or to cite them other than as "work in progress." 67 This Internet-Draft will expire on December 26, 2016. 69 Copyright Notice 71 Copyright (c) 2016 IETF Trust and the persons identified as the 72 document authors. All rights reserved. 74 This document is subject to BCP 78 and the IETF Trust's Legal 75 Provisions Relating to IETF Documents 76 (http://trustee.ietf.org/license-info) in effect on the date of 77 publication of this document. Please review these documents 78 carefully, as they describe your rights and restrictions with respect 79 to this document. Code Components extracted from this document must 80 include Simplified BSD License text as described in Section 4.e of 81 the Trust Legal Provisions and are provided without warranty as 82 described in the Simplified BSD License. 84 Table of Contents 86 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 87 2. Definitions of Terms Used in This Memo . . . . . . . . . . . 6 88 3. Flow Specifications . . . . . . . . . . . . . . . . . . . . . 6 89 4. Traffic Filtering . . . . . . . . . . . . . . . . . . . . . . 7 90 4.1. Support for other AFIs . . . . . . . . . . . . . . . . . 8 91 5. Dissemination of IPv4 FLow Specification Information . . . . 8 92 5.1. Length Encoding . . . . . . . . . . . . . . . . . . . . . 9 93 5.2. NLRI Value Encoding . . . . . . . . . . . . . . . . . . . 9 94 5.2.1. Type 1 - Destination Prefix . . . . . . . . . . . . . 12 95 5.2.2. Type 2 - Source Prefix . . . . . . . . . . . . . . . 13 96 5.2.3. Type 3 - Source Prefix . . . . . . . . . . . . . . . 13 97 5.2.4. Type 4 - Port . . . . . . . . . . . . . . . . . . . . 13 98 5.2.5. Type 5 - Destination Port . . . . . . . . . . . . . . 14 99 5.2.6. Type 6 - Destination Port . . . . . . . . . . . . . . 14 100 5.2.7. Type 7 - ICMP type . . . . . . . . . . . . . . . . . 14 101 5.2.8. Type 8 - ICMP code . . . . . . . . . . . . . . . . . 14 102 5.2.9. Type 9 - ICMP code . . . . . . . . . . . . . . . . . 15 103 5.2.10. Type 10 - Packet length . . . . . . . . . . . . . . . 15 104 5.2.11. Type 11 - DSCP (Diffserv Code Point) . . . . . . . . 15 105 5.2.12. Type 12 - Fragment . . . . . . . . . . . . . . . . . 16 106 5.2.13. Examples of Encodings . . . . . . . . . . . . . . . . 16 107 5.3. Ordering of Traffic Filtering Rules . . . . . . . . . . . 17 108 5.4. Validation Procedure . . . . . . . . . . . . . . . . . . 18 109 6. Traffic Filtering Actions . . . . . . . . . . . . . . . . . . 20 110 6.1. Traffic Rate in bytes (sub-type 0x06) . . . . . . . . . . 21 111 6.2. Traffic-action (sub-type 0x07) . . . . . . . . . . . . . 21 112 6.3. IP Redirect (sub-type 0x08) . . . . . . . . . . . . . . . 22 113 6.4. Traffic Marking (sub-type 0x09) . . . . . . . . . . . . . 22 114 6.5. Rules on Traffic Action interference . . . . . . . . . . 22 115 7. Dissemination of Traffic Filtering in BGP/MPLS VPN Networks . 23 116 7.1. Validation Procedures for BGP/MPLS VPNs . . . . . . . . . 24 117 7.2. Traffic Actions Rules . . . . . . . . . . . . . . . . . . 24 118 8. Limitations of Previous Traffic Filtering Efforts . . . . . . 24 119 8.1. Limitations in Previous DDOS Traffic Filtering Efforts . 24 120 8.2. Limitations in Previous BGP/MPLS Traffic Monitoring . . . 25 121 8.3. Limitations in BGP Flow Specification for SDN/NFV 122 Applications . . . . . . . . . . . . . . . . . . . . . . 25 123 9. Traffic Monitoring . . . . . . . . . . . . . . . . . . . . . 25 124 10. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 26 125 10.1. AFI/SAFI Definitions . . . . . . . . . . . . . . . . . . 26 126 10.2. Flow Component definitions . . . . . . . . . . . . . . . 26 127 10.3. Extended Community Flow Specification Actions . . . . . 27 128 11. Security Considerations . . . . . . . . . . . . . . . . . . . 28 129 12. Contributing authors . . . . . . . . . . . . . . . . . . . . 28 130 13. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 29 131 14. References . . . . . . . . . . . . . . . . . . . . . . . . . 29 132 14.1. Normative References . . . . . . . . . . . . . . . . . . 29 133 14.2. Informative References . . . . . . . . . . . . . . . . . 31 134 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 32 136 1. Introduction 138 Modern IP routers contain both the capability to forward traffic 139 according to IP prefixes as well as to classify, shape, rate limit, 140 filter, or redirect packets based on administratively defined 141 policies. 143 These traffic policy mechanisms allow the router to define match 144 rules that operate on multiple fields of the packet header. Actions 145 such as the ones described above can be associated with each rule. 147 The n-tuple consisting of the matching criteria defines an aggregate 148 traffic flow specification. The matching criteria can include 149 elements such as source and destination address prefixes, IP 150 protocol, and transport protocol port numbers. 152 This document defines a general procedure to encode flow 153 specification rules for aggregated traffic flows so that they can be 154 distributed as a BGP [RFC5575] NLRI. Additionally, we define the 155 required mechanisms to utilize this definition to the problem of 156 immediate concern to the authors: intra- and inter-provider 157 distribution of traffic filtering rules to filter (distributed) 158 denial-of-service (DoS) attacks. 160 By expanding routing information with flow specifications, the 161 routing system can take advantage of the ACL (Access Control List) or 162 firewall capabilities in the router's forwarding path. Flow 163 specifications can be seen as more specific routing entries to a 164 unicast prefix and are expected to depend upon the existing unicast 165 data information. 167 A flow specification received from an external autonomous system will 168 need to be validated against unicast routing before being accepted. 169 If the aggregate traffic flow defined by the unicast destination 170 prefix is forwarded to a given BGP peer, then the local system can 171 safely install more specific flow rules that may result in different 172 forwarding behavior, as requested by this system. 174 The key technology components required to address the class of 175 problems targeted by this document are: 177 1. Efficient point-to-multipoint distribution of control plane 178 information. 180 2. Inter-domain capabilities and routing policy support. 182 3. Tight integration with unicast routing, for verification 183 purposes. 185 Items 1 and 2 have already been addressed using BGP for other types 186 of control plane information. Close integration with BGP also makes 187 it feasible to specify a mechanism to automatically verify flow 188 information against unicast routing. These factors are behind the 189 choice of BGP as the carrier of flow specification information. 191 As with previous extensions to BGP, this specification makes it 192 possible to add additional information to Internet routers. These 193 are limited in terms of the maximum number of data elements they can 194 hold as well as the number of events they are able to process in a 195 given unit of time. The authors believe that, as with previous 196 extensions, service providers will be careful to keep information 197 levels below the maximum capacity of their devices. 199 In many deployments of BGP Flow Specification, the flow specification 200 information has replace existing host length route advertisements. 202 Experience with previous BGP extensions has also shown that the 203 maximum capacity of BGP speakers has been gradually increased 204 according to expected loads. Taking into account Internet unicast 205 routing as well as additional applications as they gain popularity. 207 From an operational perspective, the utilization of BGP as the 208 carrier for this information allows a network service provider to 209 reuse both internal route distribution infrastructure (e.g., route 210 reflector or confederation design) and existing external 211 relationships (e.g., inter-domain BGP sessions to a customer 212 network). 214 While it is certainly possible to address this problem using other 215 mechanisms, this solution has been utilized in deployments because of 216 the substantial advantage of being an incremental addition to already 217 deployed mechanisms. 219 In current deployments, the information distributed by the flow-spec 220 extension is originated both manually as well as automatically. The 221 latter by systems that are able to detect malicious flows. When 222 automated systems are used, care should be taken to ensure their 223 correctness as well as to limit the number and advertisement rate of 224 flow routes. 226 This specification defines required protocol extensions to address 227 most common applications of IPv4 unicast and VPNv4 unicast filtering. 228 The same mechanism can be reused and new match criteria added to 229 address similar filtering needs for other BGP address families such 230 as: 232 o IPv6 [I-D.ietf-idr-flow-spec-v6], 234 o MAC address for L2VPN [I-D.ietf-idr-flowspec-l2vpn], 236 o NV03 encapsulation [I-D.ietf-idr-flowspec-nvo3], 237 o and MPLS ([I-D.ietf-idr-flowspec-mpls-match], 238 [I-D.ietf-idr-bgp-flowspec-label]). 240 These additions to BGP Flow Specification IPv4 are included in a 241 separate documents to allow implementers the choice of implementing 242 portions of the BGP Flow specification. 244 2. Definitions of Terms Used in This Memo 246 NLRI - Network Layer Reachability Information. 248 RIB - Routing Information Base. 250 Loc-RIB - Local RIB. 252 AS - Autonomous System number. 254 VRF - Virtual Routing and Forwarding instance. 256 PE - Provider Edge router 258 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 259 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 260 document are to be interpreted as described in [RFC2119] 262 3. Flow Specifications 264 A flow specification is an n-tuple consisting of several matching 265 criteria that can be applied to IP traffic. A given IP packet is 266 said to match the defined flow if it matches all the specified 267 criteria. 269 A given flow may be associated with a set of attributes, depending on 270 the particular application; such attributes may or may not include 271 reachability information (i.e., NEXT_HOP). Well-known or AS-specific 272 community attributes can be used to encode a set of predetermined 273 actions. 275 A particular application is identified by a specific (Address Family 276 Identifier, Subsequent Address Family Identifier (AFI, SAFI)) pair 277 [RFC4760] and corresponds to a distinct set of RIBs. Those RIBs 278 should be treated independently from each other in order to assure 279 non-interference between distinct applications. 281 BGP itself treats the NLRI as an opaque key to an entry in its 282 databases. Entries that are placed in the Loc-RIB are then 283 associated with a given set of semantics, which is application 284 dependent. This is consistent with existing BGP applications. For 285 instance, IP unicast routing (AFI=1, SAFI=1) and IP multicast 286 reverse-path information (AFI=1, SAFI=2) are handled by BGP without 287 any particular semantics being associated with them until installed 288 in the Loc-RIB. 290 Standard BGP policy mechanisms, such as UPDATE filtering by NLRI 291 prefix and community matching, SHOULD apply to the Flow specification 292 defined NLRI-type. Network operators can also control propagation of 293 such routing updates by enabling or disabling the exchange of a 294 particular (AFI, SAFI) pair on a given BGP peering session. 296 4. Traffic Filtering 298 Traffic filtering policies have been traditionally considered to be 299 relatively static. Limitations of the static mechanisms caused this 300 mechanism to be designed for the three new applications of traffic 301 filtering (prevention of traffic-based, denial-of-service (DOS) 302 attacks, traffic filtering in the context of BGP/MPLS VPN service, 303 and centralized traffic control for SDN/NFV networks) requires 304 coordination among service providers and/or coordination among the AS 305 within a service provider. Section 8 has details on the limitation 306 of previous mechanisms and why BGP Flow Specification version 1 307 provides a solution for to prevent DOS and aid BGP/MPLS VPN filtering 308 rules. 310 This flow specification NLRI defined above to convey information 311 about traffic filtering rules for traffic that should be discarded or 312 handled in manner specified by a set of pre-defined actions (which 313 are defined in BGP Extended Communities). This mechanism is 314 primarily designed to allow an upstream autonomous system to perform 315 inbound filtering in their ingress routers of traffic that a given 316 downstream AS wishes to drop. 318 In order to achieve this goal, this draft specifies two application 319 specific NLRI identifiers that provide traffic filters, and a set of 320 actions encoding in BGP Extended Communities. The two application 321 specific NLRI identifiers are: 323 o IPv4 flow specification identifier (AFI=1, SAFI=133) along with 324 specific semantic rules for IPv4 routes, and 326 o BGP NLRI type (AFI=1, SAFI=134) value, which can be used to 327 propagate traffic filtering information in a BGP/MPLS VPN 328 environment. 330 Distribution of the IPv4 Flow specification is described in section 331 6, and distibution of BGP/MPLS traffic flow specification is 332 described in section 8. The traffic filtering actions are described 333 in section 7. 335 4.1. Support for other AFIs 337 Other documents shown in table 5 provide the application identifiers 338 for IPv6, L2VPN, NVO3 and MPLS. However, to provide backward 339 compatiblity with [RFC5575] documents adhering to this specification 340 do not need to support IPv6, L2VPN, NV03, and MPLS AFI/SAFIs. 342 Table 5 - AFI/SAFI values vs. application 343 +---+----+-----------+-----------------------------------+---+ 344 |AFI|SAFI|Application| Document |Req| 345 +---+----+-----------+-----------------------------------+---+ 346 | 1| 133| DDOS | this document |Yes| 347 | 1| 134| BGP/MPLS | this document | No| 348 +---+----------------+-----------------------------------+---+ 349 | 2| 133| DDOS |draft-ietf-idr-flow-spec-v6 | No| 350 | 2| 134| BGP/MPLS |draft-ietf-idr-flow-spec-v6 | No| 351 +---+----+-----------+-----------------------------------+---+ 352 | 25| 133| DDOS |draft-ietf-idr-flowspec-l2vpn | No| 353 | 25| 134| BGP/MPLS |draft-ietf-idr-flowspec-l2vpn | No| 354 +---+----+-----------+-----------------------------------+---+ 355 |TBD| 133| DDOS |draft-ietf-idr-flowspec-mpls-label | No| 356 |TBD| 134| BGP/MPLS |draft-ietf-idr-flowspec-mpls-label | No| 357 +---+----+-----------+-----------------------------------+---+ 358 |TBD| 133| DDOS |draft-ietf-idr-flowspec-nv03 | No| 359 |TBD| 134| BGP/MPLS |draft-ietf-idr-flowspec-nv03 | No| 360 +---+----+-----------+-----------------------------------+---+ 362 5. Dissemination of IPv4 FLow Specification Information 364 We define a "Flow Specification" NLRI type that may include several 365 components such as destination prefix, source prefix, protocol, 366 ports, and others (see Tables 1-4 below). This NLRI is treated as an 367 opaque bit string prefix by BGP. Each bit string identifies a key to 368 a database entry with which a set of attributes can be associated. 370 This NLRI information is encoded using MP_REACH_NLRI and 371 MP_UNREACH_NLRI attributes as defined in [RFC4760]. Whenever the 372 corresponding application does not require Next-Hop information, this 373 shall be encoded as a 0-octet length Next Hop in the MP_REACH_NLRI 374 attribute and ignored on receipt. 376 The NLRI field of the MP_REACH_NLRI and MP_UNREACH_NLRI is encoded as 377 a 1- or 2-octet NLRI length field followed by a variable-length NLRI 378 value. The NLRI length is expressed in octets. 380 +------------------------------+ 381 | length (0xnn or 0xfn nn) | 382 +------------------------------+ 383 | NLRI value (variable) | 384 +------------------------------+ 386 Figure 1: Flow-spec NLRI for IPv4 388 Implementations wishing to exchange flow specification rules MUST use 389 BGP's Capability Advertisement facility to exchange the Multiprotocol 390 Extension Capability Code (Code 1) as defined in [RFC4760]. The 391 (AFI, SAFI) pair carried in the Multiprotocol Extension Capability 392 MUST be the same as the one used to identify a particular application 393 that uses this NLRI-type. 395 5.1. Length Encoding 397 o If the NLRI length value is smaller than 240 (0xf0 hex), the 398 length field can be encoded as a single octet. 400 o Otherwise, it is encoded as an extended-length 2-octet value in 401 which the most significant nibble of the first byte is all ones. 403 In figure 1 above, values less-than 240 are encoded using two hex 404 digits (0xnn). Values above 240 are encoded using 3 hex digits 405 (0xfnnn). The highest value that can be represented with this 406 encoding is 4095. The value 241 is encoded as 0xf0f1. 408 5.2. NLRI Value Encoding 410 The Flow specification NLRI-type consists of several optional 411 subcomponents. A specific packet is considered to match the flow 412 specification when it matches the intersection (AND) of all the 413 components present in the specification. The encoding of each of the 414 NLRI components begins with a type field as listed in Table 1-4. 415 Sections 4.2.1 to 4.2.12 contain the specific encodings for the IPv4 416 IP layer and transport layer headings. Additional filters encodiings 417 are described in the following documents. 419 o IPv4 Header - this document, 421 o IPv6 Header - [I-D.ietf-idr-flow-spec-v6], 423 o MPLS label - [I-D.ietf-idr-flowspec-mpls-match], 425 o MAC Addresses - [I-D.ietf-idr-flowspec-l2vpn], and 427 o NV03 encapsulations - [I-D.ietf-idr-flowspec-nvo3]. 429 o Vendor Specific Flow Specifications (First Come First Served) 430 specifications - range is defined in this document. 432 Flow specification components must follow strict type ordering by 433 increasing numerical order. A given component type may or may not be 434 present in the specification, but if present, it MUST precede any 435 component of higher numeric type value. 437 If a given component type within a prefix in unknown, the prefix in 438 question cannot be used for traffic filtering purposes by the 439 receiver. Since a flow specification has the semantics of a logical 440 AND of all components, if a component is FALSE, by definition it 441 cannot be applied. However, for the purposes of BGP route 442 propagation, this prefix should still be transmitted since BGP route 443 distribution is independent on NLRI semantics. 445 The 540 Defines: the destination prefix to match. Prefixes are encoded as 541 in BGP UPDATE messages, a length in bits is followed by enough 542 octets to contain the prefix information. 544 5.2.2. Type 2 - Source Prefix 546 Encoding: 548 Defines the source prefix to match. 550 5.2.3. Type 3 - Source Prefix 552 Encoding: 554 Contains a set of {operator, value} pairs that are used to match 555 the IP protocol value byte in IP packets. 557 The operator byte is encoded as: 559 0 1 2 3 4 5 6 7 560 +---+---+---+---+---+---+---+---+ 561 | e | a | len | 0 |lt |gt |eq | 562 +---+---+---+---+---+---+---+---+ 564 Numerical operator 566 e - end-of-list bit. Set in the last {op, value} pair in the 567 list. 569 a - AND bit. If unset, the previous term is logically ORed with 570 the current one. If set, the operation is a logical AND. It 571 should be unset in the first operator byte of a sequence. The AND 572 operator has higher priority than OR for the purposes of 573 evaluating logical expressions. 575 len - length of the value field for this operand is given as (1 << 576 len). 578 lt - less than comparison between data and value. 580 gt - greater than comparison between data and value. 582 eq -equality between data and value 584 The bits lt, gt, and eq can be combined to produce "less or equal", 585 "greater or equal", and inequality values 587 5.2.4. Type 4 - Port 589 Encoding: 590 Defines a list of {operation, value} pairs that matches source OR 591 destination TCP/UDP ports. This list is encoded using the numeric 592 operand format defined above. Values are encoded as 1- or 2-byte 593 quantities. 595 Port, source port, and destination port components evaluate to 596 FALSE if the IP protocol field of the packet has a value other 597 than TCP or UDP, if the packet is fragmented and this is not the 598 first fragment, or if the system in unable to locate the transport 599 header. Different implementations may or may not be able to 600 decode the transport header in the presence of IP options or 601 Encapsulating Security Payload (ESP) NULL [RFC4303] encryption. 603 5.2.5. Type 5 - Destination Port 605 Encoding: 607 Defines a list of {operation, value} pairs used to match the 608 destination port of a TCP or UDP packet. Values are encoded as 1- 609 or 2-byte quantities 611 5.2.6. Type 6 - Destination Port 613 Encoding: 615 Defines a list of {operation, value} pairs used to match the 616 source port of a TCP or UDP packet. Values are encoded as 1- or 617 2-byte quantities 619 5.2.7. Type 7 - ICMP type 621 Encoding: 623 Defines a list of {operation, value} pairs used to match the type 624 field of an ICMP packet. Values are encoded using a single byte. 626 The ICMP type and code specifiers evaluate to FALSE whenever the 627 protocol value is not ICMP. 629 5.2.8. Type 8 - ICMP code 631 Encoding: 633 Defines a list of {operation, value} pairs used to match the code 634 field of an ICMP packet. Values are encoded using a single byte. 636 5.2.9. Type 9 - ICMP code 638 Encoding: 640 Bitmask values can be encoded as a 1- or 2-byte bitmask. When a 641 single byte is specified, it matches byte 13 of the TCP header 642 [RFC0793], which contains bits 8 though 15 of the 4th 32-bit word. 643 When a 2-byte encoding is used, it matches bytes 12 and 13 of the 644 TCP header with the data offset field having a "don't care" value. 646 As with port specifiers, this component evaluates to FALSE for 647 packets that are not TCP packets. 649 This type uses the bitmask operand format, which differs from the 650 numeric operator format in the lower nibble. 652 0 1 2 3 4 5 6 7 653 +---+---+---+---+---+---+---+---+ 654 | e | a | len | 0 | 0 |not| m | 655 +---+---+---+---+---+---+---+---+ 657 Bitmask format 659 e, a, len - Most significant nibble: (end-of-list bit, AND bit, and 660 length field), as defined for in the numeric operator format. 662 not - NOT bit. If set, logical negation of operation. 664 m - Match bit. If set, this is a bitwise match operation defined 665 as "(data AND value) == value"; if unset, (data AND value) 666 evaluates to TRUE if any of the bits in the value mask are set in 667 the data 669 5.2.10. Type 10 - Packet length 671 Encoding: 673 Defines match on the total IP packet length (excluding Layer 2 but 674 including IP header). Values are encoded using 1- or 2-byte 675 quantities. 677 5.2.11. Type 11 - DSCP (Diffserv Code Point) 679 Encoding: 681 Defines a list of {operation, value} pairs used to match the 6-bit 682 DSCP field [RFC2474]. Values are encoded using a single byte, 683 where the two most significant bits are zero and the six least 684 significant bits contain the DSCP value. 686 5.2.12. Type 12 - Fragment 688 Encoding: 690 Uses bitmask operand format defined above in section 5.2.9. 692 0 1 2 3 4 5 6 7 693 +---+---+---+---+---+---+---+---+ 694 | Reserved |LF |FF |IsF|DF | 695 +---+---+---+---+---+---+---+---+ 697 Bitmask values: 699 Bit 7 - Don't fragment (DF) 701 Bit 6 - Is a fragment (IsF) 703 Bit 5 - First fragment (FF) 705 Bit 4 - Last fragment (LF) 707 5.2.13. Examples of Encodings 709 An example of a flow specification encoding for: "all packets to 710 10.0.1/24 and TCP port 25". 712 +------------------+----------+----------+ 713 | destination | proto | port | 714 +------------------+----------+----------+ 715 | 0x01 18 0a 00 01 | 03 81 06 | 04 81 19 | 716 +------------------+----------+----------+ 718 Decode for protocol: 720 +-------+----------+------------------------------+ 721 | Value | | | 722 +-------+----------+------------------------------+ 723 | 0x03 | type | | 724 | 0x81 | operator | end-of-list, value size=1, = | 725 | 0x06 | value | | 726 +-------+----------+------------------------------+ 728 An example of a flow specification encoding for: "all packets to 729 10.0.1/24 from 192/8 and port {range [137, 139] or 8080}". 731 +------------------+----------+-------------------------+ 732 | destination | source | port | 733 +------------------+----------+-------------------------+ 734 | 0x01 18 0a 01 01 | 02 08 c0 | 04 03 89 45 8b 91 1f 90 | 735 +------------------+----------+-------------------------+ 737 Decode for port: 739 +--------+----------+------------------------------+ 740 | Value | | | 741 +--------+----------+------------------------------+ 742 | 0x04 | type | | 743 | 0x03 | operator | size=1, >= | 744 | 0x89 | value | 137 | 745 | 0x45 | operator | "AND", value size=1, <= | 746 | 0x8b | value | 139 | 747 | 0x91 | operator | end-of-list, value-size=2, = | 748 | 0x1f90 | value | 8080 | 749 +--------+----------+------------------------------+ 751 This constitutes an NLRI with an NLRI length of 16 octets. 753 5.3. Ordering of Traffic Filtering Rules 755 With traffic filtering rules, more than one rule may match a 756 particular traffic flow. Thus, it is necessary to define the order 757 at which rules get matched and applied to a particular traffic flow. 758 This ordering function must be such that it must not depend on the 759 arrival order of the flow specification's rules and must be constant 760 in the network. 762 The relative order of two flow specification rules is determined by 763 comparing their respective components. The algorithm starts by 764 comparing the left-most components of the rules. If the types 765 differ, the rule with lowest numeric type value has higher precedence 766 (and thus will match before) than the rule that doesn't contain that 767 component type. If the component types are the same, then a type- 768 specific comparison is performed. 770 For IP prefix values (IP destination and source prefix) precedence is 771 given to the lowest IP value of the common prefix length; if the 772 common prefix is equal, then the most specific prefix has precedence. 774 For all other component types, unless otherwise specified, the 775 comparison is performed by comparing the component data as a binary 776 string using the memcmp() function as defined by the ISO C standard. 777 For strings of different lengths, the common prefix is compared. If 778 equal, the longest string is considered to have higher precedence 779 than the shorter one. 781 Pseudocode: 783 flow_rule_cmp (a, b) 784 { 785 comp1 = next_component(a); 786 comp2 = next_component(b); 787 while (comp1 || comp2) { 788 // component_type returns infinity on end-of-list 789 if (component_type(comp1) < component_type(comp2)) { 790 return A_HAS_PRECEDENCE; 791 } 792 if (component_type(comp1) > component_type(comp2)) { 793 return B_HAS_PRECEDENCE; 794 } 796 if (component_type(comp1) == IP_DESTINATION || IP_SOURCE) { 797 common = MIN(prefix_length(comp1), prefix_length(comp2)); 798 cmp = prefix_compare(comp1, comp2, common); 799 // not equal, lowest value has precedence 800 // equal, longest match has precedence 801 } else { 802 common = 803 MIN(component_length(comp1), component_length(comp2)); 804 cmp = memcmp(data(comp1), data(comp2), common); 805 // not equal, lowest value has precedence 806 // equal, longest string has precedence 807 } 808 } 810 return EQUAL; 811 } 813 5.4. Validation Procedure 815 Flow specifications received from a BGP peer and that are accepted in 816 the respective Adj-RIB-In are used as input to the route selection 817 process. Although the forwarding attributes of two routes for the 818 same flow specification prefix may be the same, BGP is still required 819 to perform its path selection algorithm in order to select the 820 correct set of attributes to advertise. 822 The first step of the BGP Route Selection procedure (Section 9.1.2 of 823 [RFC4271] is to exclude from the selection procedure routes that are 824 considered non-feasible. In the context of IP routing information, 825 this step is used to validate that the NEXT_HOP attribute of a given 826 route is resolvable. 828 The concept can be extended, in the case of flow specification NLRI, 829 to allow other validation procedures. 831 A flow specification NLRI must be validated such that it is 832 considered feasible if and only if: 834 a) The originator of the flow specification matches the originator 835 of the best-match unicast route for the destination prefix 836 embedded in the flow specification. 838 b) There are no more specific unicast routes, when compared with 839 the flow destination prefix, that have been received from a 840 different neighboring AS than the best-match unicast route, which 841 has been determined in step a). 843 By originator of a BGP route, we mean either the BGP originator path 844 attribute, as used by route reflection, or the transport address of 845 the BGP peer, if this path attribute is not present. 847 The underlying concept is that the neighboring AS that advertises the 848 best unicast route for a destination is allowed to advertise flow- 849 spec information that conveys a more or equally specific destination 850 prefix. Thus, as long as there are no more specific unicast routes, 851 received from a different neighboring AS, which would be affected by 852 that filtering rule. 854 The neighboring AS is the immediate destination of the traffic 855 described by the flow specification. If it requests these flows to 856 be dropped, that request can be honored without concern that it 857 represents a denial of service in itself. Supposedly, the traffic is 858 being dropped by the downstream autonomous system, and there is no 859 added value in carrying the traffic to it. 861 BGP implementations MUST also enforce that the AS_PATH attribute of a 862 route received via the External Border Gateway Protocol (eBGP) 863 contains the neighboring AS in the left-most position of the AS_PATH 864 attribute. While this rule is optional in the BGP specification, it 865 becomes necessary to enforce it for security reasons. 867 6. Traffic Filtering Actions 869 This specification defines a minimum set of filtering actions that it 870 standardizes as BGP extended community values [RFC4360]. This is not 871 meant to be an inclusive list of all the possible actions, but only a 872 subset that can be interpreted consistently across the network. 873 Additional actions can be defined as either requiring standards or as 874 vendor specific. 876 Implementations SHOULD provide mechanisms that map an arbitrary BGP 877 community value (normal or extended) to filtering actions that 878 require different mappings in different systems in the network. For 879 instance, providing packets with a worse-than-best-effort, per-hop 880 behavior is a functionality that is likely to be implemented 881 differently in different systems and for which no standard behavior 882 is currently known. Rather than attempting to define it here, this 883 can be accomplished by mapping a user-defined community value to 884 platform-/network-specific behavior via user configuration. 886 The default action for a traffic filtering flow specification is to 887 accept IP traffic that matches that particular rule. 889 This document defines the following extended communities values shown 890 in table X in the form 0x8xnn where nn indicates the sub-type. 892 Table 5 - Traffic Action Extended Communities 893 Defined in this document 895 +--------+-----------------------+---------------------------+ 896 | type | extended community | encoding | 897 +--------+-----------------------+---------------------------+ 898 | 0x8006 | traffic-rate in bytes | 2-byte ASN, 4-byte float | 899 | 0x8007 | traffic-action | bitmask | 900 | 0x8008 | redirect IPv4 AS-2byte| 2-octet ASN, 4-octet value| 901 | 0x8108 | redirect IPv4 address | 4-octet IPv4 Address, | 902 | | | 2-octet Value | 903 | 0x8208 | redirect IPv4 AS-4byte| 4-octet AS, 2-octet Value | 904 | 0x8009 | traffic-marking | DSCP value | 905 +--------+-----------------------+---------------------------+ 907 Encodings for these extended communities are described below. 909 Some traffic action communities may interfere with each other. 910 Section x.x of this specification provides rules for handling 911 interference between specific types of traffic actions, and error 912 handling based on [RFC7606] in section. Each definition of a traffic 913 action MUST specify any interface with any other traffic actions, any 914 impact on flow specification process, and error handling per 915 [RFC7606]. 917 The traffic actions are processed in ascending order of the sub-type 918 found in the BGP Extended Communities. 920 6.1. Traffic Rate in bytes (sub-type 0x06) 922 The traffic-rate extended community is a non- transitive extended 923 community across the autonomous-system boundary and uses following 924 extended community encoding: 926 The first two octets carry the 2-octet id, which can be assigned from 927 a 2-byte AS number. When a 4-byte AS number is locally present, the 928 2 least significant bytes of such an AS number can be used. This 929 value is purely informational and should not be interpreted by the 930 implementation. 932 The remaining 4 octets carry the maximum rate information in IEEE 933 floating point [IEEE.754.1985] format, units being bytes per second. 934 A traffic-rate of 0 should result on all traffic for the particular 935 flow to be discarded. 937 Interfers with: Traffic Rate in packets. Process traffic rate in 938 bytes (sub-type 0x06) action before traffic rate action (sub-type 939 TBD). 941 6.2. Traffic-action (sub-type 0x07) 943 The traffic-action extended community consists of 6 bytes of which 944 only the 2 least significant bits of the 6th byte (from left to 945 right) are currently defined. 947 40 41 42 43 44 45 46 47 948 +---+---+---+---+---+---+---+---+ 949 | reserved | S | T | 950 +---+---+---+---+---+---+---+---+ 952 where S and T are defined as: 954 o T: Terminal Action (bit 47): When this bit is set, the traffic 955 filtering engine will apply any subsequent filtering rules (as 956 defined by the ordering procedure). If not set, the evaluation of 957 the traffic filter stops when this rule is applied. 959 o S:Sample (bit 46): Enables traffic sampling and logging for this 960 flow specification. 962 Interfers with: No other BGP Flow Specification traffic action in 963 this document. 965 6.3. IP Redirect (sub-type 0x08) 967 The redirect extended community allows the traffic to be redirected 968 to a VRF routing instance that lists the specified route-target in 969 its import policy. If several local instances match this criteria, 970 the choice between them is a local matter (for example, the instance 971 with the lowest Route Distinguisher value can be elected). This 972 extended community uses the same encoding as the Route Target 973 extended community [RFC4360]. 975 It should be noted that the low-order nibble of the Redirect's Type 976 field corresponds to the Route Target Extended Community format field 977 (Type). (See Sections 3.1, 3.2, and 4 of [RFC4360] plus Section 2 of 978 [RFC5668].) The low-order octet (Sub-Type) of the Redirect Extended 979 Community remains 0x08 for all three encodings of the BGP Extended 980 Communities (AS 2-byte, AS 4-byte, and IPv4 address). 982 Interfers with: All other redirect functions. All redirect functions 983 are mutually exclusive. If this redirect function exists, then no 984 other redirect functions can be processed. 986 6.4. Traffic Marking (sub-type 0x09) 988 The traffic marking extended community instructs a system to modify 989 the DSCP bits of a transiting IP packet to the corresponding value. 990 This extended community is encoded as a sequence of 5 zero bytes 991 followed by the DSCP value encoded in the 6 least significant bits of 992 6th byte. 994 Interfers with: No other action in this document. 996 6.5. Rules on Traffic Action interference 998 The following traffic Actions may interfere with each other: 1000 o redirect actions, 1002 o traffic rate actions, and 1004 o encapsulation actions. 1006 This specification has the following rules regaarding multiple 1007 traffic actions to prevent the interference: 1009 1. All redirect actions are mutually exclusive. Presence of more 1010 than one results in no redirect. 1012 2. If multiple rate actions are present, these are applied in 1013 ascending order of the sub-type. 1015 3. Some actions are unique, and may operate independently. For 1016 example, an MPLS push/pop action is unique. 1018 4. Each additional flow specification Action must specify: 1020 * whether it is a redirect or rate action, 1022 * whether the action is unique or if it interfers with other 1023 actions, 1025 * If the action interfers with other actions, the handling must 1026 be specified if both the action and other interfering actions 1027 exist are associated with a Flow specification NLRI. 1029 7. Dissemination of Traffic Filtering in BGP/MPLS VPN Networks 1031 Provider-based Layer 3 VPN networks, such as the ones using a BGP/ 1032 MPLS IP VPN [RFC4364] control plane, have different traffic filtering 1033 requirements than Internet service providers. This document proposes 1034 an additional BGP NLRI type (AFI=1, SAFI=134) value, which can be 1035 used to propagate traffic filtering information in a BGP/MPLS VPN 1036 environment. 1038 The NLRI format for this address family consists of a fixed-length 1039 Route Distinguisher field (8 bytes) followed by a flow specification, 1040 following the encoding defined above in section x of this document. 1041 The NLRI length field shall include both the 8 bytes of the Route 1042 Distinguisher as well as the subsequent flow specification. 1044 +------------------------------+ 1045 | length (0xnn or 0xfn nn) | 1046 +------------------------------+ 1047 | Route Distinguisher (8 bytes)| 1048 +------------------------------+ 1049 | NLRI value (variable) | 1050 +------------------------------+ 1052 Figure 2: Flow-spec NLRI for MPLS 1054 Propagation of this NLRI is controlled by matching Route Target 1055 extended communities associated with the BGP path advertisement with 1056 the VRF import policy, using the same mechanism as described in "BGP/ 1057 MPLS IP VPNs" [RFC4364]. 1059 Flow specification rules received via this NLRI apply only to traffic 1060 that belongs to the VRF(s) in which it is imported. By default, 1061 traffic received from a remote PE is switched via an MPLS forwarding 1062 decision and is not subject to filtering. 1064 Contrary to the behavior specified for the non-VPN NLRI, flow rules 1065 are accepted by default, when received from remote PE routers. 1067 7.1. Validation Procedures for BGP/MPLS VPNs 1069 The validation procedures are the same as for IPv4. 1071 7.2. Traffic Actions Rules 1073 The traffic action rules are the same as for IPv4. 1075 8. Limitations of Previous Traffic Filtering Efforts 1077 8.1. Limitations in Previous DDOS Traffic Filtering Efforts 1079 The popularity of traffic-based, denial-of-service (DoS) attacks, 1080 which often requires the network operator to be able to use traffic 1081 filters for detection and mitigation, brings with it requirements 1082 that are not fully satisfied by existing tools. 1084 Increasingly, DoS mitigation requires coordination among several 1085 service providers in order to be able to identify traffic source(s) 1086 and because the volumes of traffic may be such that they will 1087 otherwise significantly affect the performance of the network. 1089 Several techniques are currently used to control traffic filtering of 1090 DoS attacks. Among those, one of the most common is to inject 1091 unicast route advertisements corresponding to a destination prefix 1092 being attacked. One variant of this technique marks such route 1093 advertisements with a community that gets translated into a discard 1094 Next-Hop by the receiving router. Other variants attract traffic to 1095 a particular node that serves as a deterministic drop point. 1097 Using unicast routing advertisements to distribute traffic filtering 1098 information has the advantage of using the existing infrastructure 1099 and inter-AS communication channels. This can allow, for instance, a 1100 service provider to accept filtering requests from customers for 1101 address space they own. 1103 There are several drawbacks, however. An issue that is immediately 1104 apparent is the granularity of filtering control: only destination 1105 prefixes may be specified. Another area of concern is the fact that 1106 filtering information is intermingled with routing information. 1108 The mechanism defined in this document is designed to address these 1109 limitations. We use the flow specification NLRI defined above to 1110 convey information about traffic filtering rules for traffic that 1111 should be discarded. 1113 8.2. Limitations in Previous BGP/MPLS Traffic Monitoring 1115 Provider-based Layer 3 VPN networks, such as the ones using a BGP/ 1116 MPLS IP VPN [RFC4364] control plane, have different traffic filtering 1117 requirements than Internet service providers. 1119 In these environments, the VPN customer network often has traffic 1120 filtering capabilities towards their external network connections 1121 (e.g., firewall facing public network connection). Less common is 1122 the presence of traffic filtering capabilities between different VPN 1123 attachment sites. In an any-to-any connectivity model, which is the 1124 default, this means that site-to-site traffic is unfiltered. 1126 In circumstances where a security threat does get propagated inside 1127 the VPN customer network, there may not be readily available 1128 mechanisms to provide mitigation via traffic filter. 1130 The BGP Flow Specification version 1 addresses these limitations. 1132 8.3. Limitations in BGP Flow Specification for SDN/NFV Applications 1134 The SDN/NFV applications which use centralized control of network 1135 traffic via dynamic distribution of traffic filters can utilize the 1136 BGP Flow Specification version 1 described in this draft with a fixed 1137 order to traffic filter matches. However, for control of large 1138 amounts of data a user-defined order to traffic matches and actions 1139 may be required. 1141 9. Traffic Monitoring 1143 Traffic filtering applications require monitoring and traffic 1144 statistics facilities. While this is an implementation-specific 1145 choice, implementations SHOULD provide: 1147 o A mechanism to log the packet header of filtered traffic. 1149 o A mechanism to count the number of matches for a given flow 1150 specification rule. 1152 10. IANA Considerations 1154 This section complies with [RFC7153] 1156 10.1. AFI/SAFI Definitions 1158 For the purpose of this work, IANA has allocated values for two 1159 SAFIs: SAFI 133 for IPv4 dissemination of flow specification rules 1160 and SAFI 134 for VPNv4 dissemination of flow specification rules. 1162 10.2. Flow Component definitions 1164 A flow specification consists of a sequence of flow components, which 1165 are identified by a an 8-bit component type. Types must be assigned 1166 and interpreted uniquely. The current specification defines types 1 1167 though 12, with the value 0 being reserved. 1169 IANA created and maintains a new registry entitled: "Flow Spec 1170 Component Types". The following component types have been 1171 registered: 1173 Type 1 - Destination Prefix 1175 Type 2 - Source Prefix 1177 Type 3 - IP Protocol 1179 Type 4 - Port 1181 Type 5 - Destination port 1183 Type 6 - Source port 1185 Type 7 - ICMP type 1187 Type 8 - ICMP code 1189 Type 9 - TCP flags 1191 Type 10 - Packet length 1193 Type 11 - DSCP 1195 Type 12 - Fragment 1197 In order to manage the limited number space and accommodate several 1198 usages, the following policies defined by RFC 5226 [RFC5226] are 1199 used: 1201 +--------------+-------------------------------+ 1202 | Range | Policy | 1203 +--------------+-------------------------------+ 1204 | 0 | Invalid value | 1205 | [1 .. 12] | Defined by this specification | 1206 | [13 .. 127] | Specification Required | 1207 | [128 .. 255] | First Come First Served | 1208 +--------------+-------------------------------+ 1210 The specification of a particular "flow component type" must clearly 1211 identify what the criteria used to match packets forwarded by the 1212 router is. This criteria should be meaningful across router hops and 1213 not depend on values that change hop-by-hop such as TTL or Layer 2 1214 encapsulation. 1216 The "traffic-action" extended community defined in this document has 1217 46 unused bits, which can be used to convey additional meaning. IANA 1218 created and maintains a new registry entitled: "Traffic Action 1219 Fields". These values should be assigned via IETF Review rules only. 1220 The following traffic-action fields have been allocated: 1222 47 Terminal Action 1224 46 Sample 1226 0-45 Unassigned 1228 10.3. Extended Community Flow Specification Actions 1230 The Extended Community FLow Specification Action types consists of 1231 two parts: BGP Transitive Extended Community types and a set of sub- 1232 types. 1234 IANA has updated the following "BGP Transitive Extended Community 1235 Types" registries to contain the values listed below: 1237 0x80 - Generic Transitive Experimental Use Extended Community Part 1238 1 (Sub-Types are defined in the "Generic Transitive Experimental 1239 Extended Community Part 1 Sub-Types" Registry) 1241 0x81 - Generic Transitive Experimental Use Extended Community Part 1242 2 (Sub-Types are defined in the "Generic Transitive Experimental 1243 Extended Community Part 2 Sub-Types" Registry) 1245 0x82 - Generic Transitive Experimental Use Extended Community Part 1246 3 (Sub-Types are defined in the "Generic Transitive Experimental 1247 Use Extended Community Part 3 Sub-Types" Registry) 1248 RANGE REGISTRATION PROCEDURE 1249 0x00-0xbf First Come First Served 1250 0xc0-0xff IETF Review 1252 SUB-TYPE VALUE NAME REFERENCE 1253 0x00-0x05 unassigned 1254 0x06 traffic-rate [this document] 1255 0x07 traffic-action [this document] 1256 0x08 Flow spec redirect IPv4 [RFC5575] [RFC7674] 1257 [this document] 1258 0x09 traffic-marking [this document] 1259 0x10-0xff Unassigned [this document] 1261 11. Security Considerations 1263 Inter-provider routing is based on a web of trust. Neighboring 1264 autonomous systems are trusted to advertise valid reachability 1265 information. If this trust model is violated, a neighboring 1266 autonomous system may cause a denial-of-service attack by advertising 1267 reachability information for a given prefix for which it does not 1268 provide service. 1270 As long as traffic filtering rules are restricted to match the 1271 corresponding unicast routing paths for the relevant prefixes, the 1272 security characteristics of this proposal are equivalent to the 1273 existing security properties of BGP unicast routing. 1275 Where it is not the case, this would open the door to further denial- 1276 of-service attacks. 1278 Enabling firewall-like capabilities in routers without centralized 1279 management could make certain failures harder to diagnose. For 1280 example, it is possible to allow TCP packets to pass between a pair 1281 of addresses but not ICMP packets. It is also possible to permit 1282 packets smaller than 900 or greater than 1000 bytes to pass between a 1283 pair of addresses, but not packets whose length is in the range 900- 1284 1000. Such behavior may be confusing and these capabilities should 1285 be used with care whether manually configured or coordinated through 1286 the protocol extensions described in this document. 1288 12. Contributing authors 1290 Pedro Marques and Nischal Sheth were authors on [RFC5575], and 1291 therefore are contributing authors on this document. 1293 13. Acknowledgements 1295 The authors would like to thank Yakov Rekhter, Dennis Ferguson, Chris 1296 Morrow, Charlie Kaufman, and David Smith for their comments for the 1297 comments on the original [RFC5575]. Chaitanya Kodeboyina helped 1298 design the flow validation procedure; and Steven Lin and Jim Washburn 1299 ironed out all the details necessary to produce a working 1300 implementation in the original [RFC5575]. 1302 Additional acknowledgements for this document will be included here. 1304 14. References 1306 14.1. Normative References 1308 [RFC0793] Postel, J., "Transmission Control Protocol", STD 7, 1309 RFC 793, DOI 10.17487/RFC0793, September 1981, 1310 . 1312 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 1313 Requirement Levels", BCP 14, RFC 2119, 1314 DOI 10.17487/RFC2119, March 1997, 1315 . 1317 [RFC2474] Nichols, K., Blake, S., Baker, F., and D. Black, 1318 "Definition of the Differentiated Services Field (DS 1319 Field) in the IPv4 and IPv6 Headers", RFC 2474, 1320 DOI 10.17487/RFC2474, December 1998, 1321 . 1323 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 1324 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 1325 DOI 10.17487/RFC4271, January 2006, 1326 . 1328 [RFC4360] Sangli, S., Tappan, D., and Y. Rekhter, "BGP Extended 1329 Communities Attribute", RFC 4360, DOI 10.17487/RFC4360, 1330 February 2006, . 1332 [RFC4364] Rosen, E. and Y. Rekhter, "BGP/MPLS IP Virtual Private 1333 Networks (VPNs)", RFC 4364, DOI 10.17487/RFC4364, February 1334 2006, . 1336 [RFC4760] Bates, T., Chandra, R., Katz, D., and Y. Rekhter, 1337 "Multiprotocol Extensions for BGP-4", RFC 4760, 1338 DOI 10.17487/RFC4760, January 2007, 1339 . 1341 [RFC4761] Kompella, K., Ed. and Y. Rekhter, Ed., "Virtual Private 1342 LAN Service (VPLS) Using BGP for Auto-Discovery and 1343 Signaling", RFC 4761, DOI 10.17487/RFC4761, January 2007, 1344 . 1346 [RFC4762] Lasserre, M., Ed. and V. Kompella, Ed., "Virtual Private 1347 LAN Service (VPLS) Using Label Distribution Protocol (LDP) 1348 Signaling", RFC 4762, DOI 10.17487/RFC4762, January 2007, 1349 . 1351 [RFC5226] Narten, T. and H. Alvestrand, "Guidelines for Writing an 1352 IANA Considerations Section in RFCs", BCP 26, RFC 5226, 1353 DOI 10.17487/RFC5226, May 2008, 1354 . 1356 [RFC5575] Marques, P., Sheth, N., Raszuk, R., Greene, B., Mauch, J., 1357 and D. McPherson, "Dissemination of Flow Specification 1358 Rules", RFC 5575, DOI 10.17487/RFC5575, August 2009, 1359 . 1361 [RFC5668] Rekhter, Y., Sangli, S., and D. Tappan, "4-Octet AS 1362 Specific BGP Extended Community", RFC 5668, 1363 DOI 10.17487/RFC5668, October 2009, 1364 . 1366 [RFC6241] Enns, R., Ed., Bjorklund, M., Ed., Schoenwaelder, J., Ed., 1367 and A. Bierman, Ed., "Network Configuration Protocol 1368 (NETCONF)", RFC 6241, DOI 10.17487/RFC6241, June 2011, 1369 . 1371 [RFC6482] Lepinski, M., Kent, S., and D. Kong, "A Profile for Route 1372 Origin Authorizations (ROAs)", RFC 6482, 1373 DOI 10.17487/RFC6482, February 2012, 1374 . 1376 [RFC7153] Rosen, E. and Y. Rekhter, "IANA Registries for BGP 1377 Extended Communities", RFC 7153, DOI 10.17487/RFC7153, 1378 March 2014, . 1380 [RFC7223] Bjorklund, M., "A YANG Data Model for Interface 1381 Management", RFC 7223, DOI 10.17487/RFC7223, May 2014, 1382 . 1384 [RFC7606] Chen, E., Ed., Scudder, J., Ed., Mohapatra, P., and K. 1385 Patel, "Revised Error Handling for BGP UPDATE Messages", 1386 RFC 7606, DOI 10.17487/RFC7606, August 2015, 1387 . 1389 [RFC7674] Haas, J., Ed., "Clarification of the Flowspec Redirect 1390 Extended Community", RFC 7674, DOI 10.17487/RFC7674, 1391 October 2015, . 1393 14.2. Informative References 1395 [I-D.ietf-idr-bgp-flowspec-label] 1396 liangqiandeng, l., Hares, S., You, J., Raszuk, R., and d. 1397 danma@cisco.com, "Carrying Label Information for BGP 1398 FlowSpec", draft-ietf-idr-bgp-flowspec-label-00 (work in 1399 progress), June 2016. 1401 [I-D.ietf-idr-bgp-flowspec-oid] 1402 Uttaro, J., Filsfils, C., Smith, D., Alcaide, J., and P. 1403 Mohapatra, "Revised Validation Procedure for BGP Flow 1404 Specifications", draft-ietf-idr-bgp-flowspec-oid-03 (work 1405 in progress), March 2016. 1407 [I-D.ietf-idr-flow-spec-v6] 1408 McPherson, D., Raszuk, R., Pithawala, B., 1409 akarch@cisco.com, a., and S. Hares, "Dissemination of Flow 1410 Specification Rules for IPv6", draft-ietf-idr-flow-spec- 1411 v6-07 (work in progress), March 2016. 1413 [I-D.ietf-idr-flowspec-interfaceset] 1414 Litkowski, S., Simpson, A., Patel, K., and J. Haas, 1415 "Applying BGP flowspec rules on a specific interface set", 1416 draft-ietf-idr-flowspec-interfaceset-01 (work in 1417 progress), June 2016. 1419 [I-D.ietf-idr-flowspec-l2vpn] 1420 Weiguo, H., liangqiandeng, l., Litkowski, S., and S. 1421 Zhuang, "Dissemination of Flow Specification Rules for L2 1422 VPN", draft-ietf-idr-flowspec-l2vpn-04 (work in progress), 1423 May 2016. 1425 [I-D.ietf-idr-flowspec-mpls-match] 1426 Yong, L., Hares, S., liangqiandeng, l., and J. You, "BGP 1427 Flow Specification Filter for MPLS Label", draft-ietf-idr- 1428 flowspec-mpls-match-00 (work in progress), May 2016. 1430 [I-D.ietf-idr-flowspec-nvo3] 1431 Weiguo, H., Zhuang, S., Li, Z., and R. Gu, "Dissemination 1432 of Flow Specification Rules for NVO3", draft-ietf-idr- 1433 flowspec-nvo3-00 (work in progress), May 2016. 1435 [I-D.ietf-idr-flowspec-packet-rate] 1436 Eddy, W., Dailey, J., and G. Clark, "BGP Flow 1437 Specification Packet-Rate Action", draft-ietf-idr- 1438 flowspec-packet-rate-00 (work in progress), June 2016. 1440 [I-D.ietf-idr-wide-bgp-communities] 1441 Raszuk, R., Haas, J., Lange, A., Amante, S., Decraene, B., 1442 Jakma, P., and R. Steenbergen, "Wide BGP Communities 1443 Attribute", draft-ietf-idr-wide-bgp-communities-02 (work 1444 in progress), May 2016. 1446 [RFC4303] Kent, S., "IP Encapsulating Security Payload (ESP)", 1447 RFC 4303, DOI 10.17487/RFC4303, December 2005, 1448 . 1450 [RFC6074] Rosen, E., Davie, B., Radoaca, V., and W. Luo, 1451 "Provisioning, Auto-Discovery, and Signaling in Layer 2 1452 Virtual Private Networks (L2VPNs)", RFC 6074, 1453 DOI 10.17487/RFC6074, January 2011, 1454 . 1456 [RFC6483] Huston, G. and G. Michaelson, "Validation of Route 1457 Origination Using the Resource Certificate Public Key 1458 Infrastructure (PKI) and Route Origin Authorizations 1459 (ROAs)", RFC 6483, DOI 10.17487/RFC6483, February 2012, 1460 . 1462 Authors' Addresses 1464 Robert Raszuk 1465 Bloomberg LP 1466 731 Lexington Ave 1467 New York City, NY 10022 1468 USA 1470 Email: robert@raszuk.net 1472 Danny McPherson 1473 Verisign 1474 US 1476 Email: dmcpherson@verisign.com 1477 Jared Mauch 1478 NTT America 1479 101 Park Ave 1480 41st Floor 1481 New York, NY 10178 1482 US 1484 Email: jmauch@us.ntt.net 1486 Barry Greene 1487 Juniper Networks 1488 1194 N. Mathilda Ave. 1489 Sunnyvale, CA 94089 1490 US 1492 Email: bgreene@juniper.net 1494 Susan Hares 1495 Huawei 1496 7453 Hickory Hill 1497 Saline, MI 48176 1498 USA 1500 Email: shares@ndzh.com