| 1 |
<!DOCTYPE html> |
| 2 |
<html lang="en" class="RFC"> |
| 3 |
<head> |
| 4 |
<meta charset="utf-8"> |
| 5 |
<meta content="Common,Latin" name="scripts"> |
| 6 |
<meta content="initial-scale=1.0" name="viewport"> |
| 7 |
<title>RFC 9117: Revised Validation Procedure for BGP Flow Specifications</title> |
| 8 |
<meta content="James Uttaro" name="author"> |
| 9 |
<meta content="Juan Alcaide" name="author"> |
| 10 |
<meta content="Clarence Filsfils" name="author"> |
| 11 |
<meta content="David Smith" name="author"> |
| 12 |
<meta content="Pradosh Mohapatra" name="author"> |
| 13 |
<meta content=" |
| 14 |
|
| 15 |
This document describes a modification to the validation procedure defined |
| 16 |
for the dissemination of BGP Flow Specifications. The dissemination of BGP |
| 17 |
Flow Specifications as specified in RFC 8955 requires that the originator |
| 18 |
of the Flow Specification match the originator of the best-match unicast |
| 19 |
route for the destination prefix embedded in the Flow Specification. For an |
| 20 |
Internal Border Gateway Protocol (iBGP) received route, the originator is |
| 21 |
typically a border router within the same autonomous system (AS). The |
| 22 |
objective is to allow only BGP speakers within the data forwarding path to |
| 23 |
originate BGP Flow Specifications. Sometimes it is desirable to originate |
| 24 |
the BGP Flow Specification from any place within the autonomous system |
| 25 |
itself, for example, from a centralized BGP route controller. However, the |
| 26 |
validation procedure described in RFC 8955 will fail in this scenario. The modification |
| 27 |
proposed herein relaxes the validation rule to enable Flow Specifications |
| 28 |
to be originated within the same autonomous system as the BGP speaker |
| 29 |
performing the validation. Additionally, this document revises the AS_PATH |
| 30 |
validation rules so Flow Specifications received from an External Border |
| 31 |
Gateway Protocol (eBGP) peer can be validated when such a peer is a BGP |
| 32 |
route server. |
| 33 |
|
| 34 |
|
| 35 |
|
| 36 |
This document updates the validation procedure in RFC 8955. |
| 37 |
|
| 38 |
" name="description"> |
| 39 |
<meta content="xml2rfc 3.9.1" name="generator"> |
| 40 |
<meta content="BGP flowspec" name="keyword"> |
| 41 |
<meta content="9117" name="rfc.number"> |
| 42 |
<!-- Generator version information: |
| 43 |
xml2rfc 3.9.1 |
| 44 |
Python 3.6.10 |
| 45 |
appdirs 1.4.4 |
| 46 |
ConfigArgParse 1.2.3 |
| 47 |
google-i18n-address 2.3.5 |
| 48 |
html5lib 1.0.1 |
| 49 |
intervaltree 3.0.2 |
| 50 |
Jinja2 2.11.2 |
| 51 |
kitchen 1.2.6 |
| 52 |
lxml 4.4.2 |
| 53 |
pycairo 1.19.0 |
| 54 |
pycountry 19.8.18 |
| 55 |
pyflakes 2.1.1 |
| 56 |
PyYAML 5.3.1 |
| 57 |
requests 2.22.0 |
| 58 |
setuptools 40.6.2 |
| 59 |
six 1.14.0 |
| 60 |
WeasyPrint 51 |
| 61 |
--> |
| 62 |
<link href="rfc9117.xml" rel="alternate" type="application/rfc+xml"> |
| 63 |
<link href="#copyright" rel="license"> |
| 64 |
<style type="text/css">/* |
| 65 |
|
| 66 |
NOTE: Changes at the bottom of this file overrides some earlier settings. |
| 67 |
|
| 68 |
Once the style has stabilized and has been adopted as an official RFC style, |
| 69 |
this can be consolidated so that style settings occur only in one place, but |
| 70 |
for now the contents of this file consists first of the initial CSS work as |
| 71 |
provided to the RFC Formatter (xml2rfc) work, followed by itemized and |
| 72 |
commented changes found necssary during the development of the v3 |
| 73 |
formatters. |
| 74 |
|
| 75 |
*/ |
| 76 |
|
| 77 |
/* fonts */ |
| 78 |
@import url('https://fonts.googleapis.com/css?family=Noto+Sans'); /* Sans-serif */ |
| 79 |
@import url('https://fonts.googleapis.com/css?family=Noto+Serif'); /* Serif (print) */ |
| 80 |
@import url('https://fonts.googleapis.com/css?family=Roboto+Mono'); /* Monospace */ |
| 81 |
|
| 82 |
@viewport { |
| 83 |
zoom: 1.0; |
| 84 |
width: extend-to-zoom; |
| 85 |
} |
| 86 |
@-ms-viewport { |
| 87 |
width: extend-to-zoom; |
| 88 |
zoom: 1.0; |
| 89 |
} |
| 90 |
/* general and mobile first */ |
| 91 |
html { |
| 92 |
} |
| 93 |
body { |
| 94 |
max-width: 90%; |
| 95 |
margin: 1.5em auto; |
| 96 |
color: #222; |
| 97 |
background-color: #fff; |
| 98 |
font-size: 14px; |
| 99 |
font-family: 'Noto Sans', Arial, Helvetica, sans-serif; |
| 100 |
line-height: 1.6; |
| 101 |
scroll-behavior: smooth; |
| 102 |
} |
| 103 |
.ears { |
| 104 |
display: none; |
| 105 |
} |
| 106 |
|
| 107 |
/* headings */ |
| 108 |
#title, h1, h2, h3, h4, h5, h6 { |
| 109 |
margin: 1em 0 0.5em; |
| 110 |
font-weight: bold; |
| 111 |
line-height: 1.3; |
| 112 |
} |
| 113 |
#title { |
| 114 |
clear: both; |
| 115 |
border-bottom: 1px solid #ddd; |
| 116 |
margin: 0 0 0.5em 0; |
| 117 |
padding: 1em 0 0.5em; |
| 118 |
} |
| 119 |
.author { |
| 120 |
padding-bottom: 4px; |
| 121 |
} |
| 122 |
h1 { |
| 123 |
font-size: 26px; |
| 124 |
margin: 1em 0; |
| 125 |
} |
| 126 |
h2 { |
| 127 |
font-size: 22px; |
| 128 |
margin-top: -20px; /* provide offset for in-page anchors */ |
| 129 |
padding-top: 33px; |
| 130 |
} |
| 131 |
h3 { |
| 132 |
font-size: 18px; |
| 133 |
margin-top: -36px; /* provide offset for in-page anchors */ |
| 134 |
padding-top: 42px; |
| 135 |
} |
| 136 |
h4 { |
| 137 |
font-size: 16px; |
| 138 |
margin-top: -36px; /* provide offset for in-page anchors */ |
| 139 |
padding-top: 42px; |
| 140 |
} |
| 141 |
h5, h6 { |
| 142 |
font-size: 14px; |
| 143 |
} |
| 144 |
#n-copyright-notice { |
| 145 |
border-bottom: 1px solid #ddd; |
| 146 |
padding-bottom: 1em; |
| 147 |
margin-bottom: 1em; |
| 148 |
} |
| 149 |
/* general structure */ |
| 150 |
p { |
| 151 |
padding: 0; |
| 152 |
margin: 0 0 1em 0; |
| 153 |
text-align: left; |
| 154 |
} |
| 155 |
div, span { |
| 156 |
position: relative; |
| 157 |
} |
| 158 |
div { |
| 159 |
margin: 0; |
| 160 |
} |
| 161 |
.alignRight.art-text { |
| 162 |
background-color: #f9f9f9; |
| 163 |
border: 1px solid #eee; |
| 164 |
border-radius: 3px; |
| 165 |
padding: 1em 1em 0; |
| 166 |
margin-bottom: 1.5em; |
| 167 |
} |
| 168 |
.alignRight.art-text pre { |
| 169 |
padding: 0; |
| 170 |
} |
| 171 |
.alignRight { |
| 172 |
margin: 1em 0; |
| 173 |
} |
| 174 |
.alignRight > *:first-child { |
| 175 |
border: none; |
| 176 |
margin: 0; |
| 177 |
float: right; |
| 178 |
clear: both; |
| 179 |
} |
| 180 |
.alignRight > *:nth-child(2) { |
| 181 |
clear: both; |
| 182 |
display: block; |
| 183 |
border: none; |
| 184 |
} |
| 185 |
svg { |
| 186 |
display: block; |
| 187 |
} |
| 188 |
.alignCenter.art-text { |
| 189 |
background-color: #f9f9f9; |
| 190 |
border: 1px solid #eee; |
| 191 |
border-radius: 3px; |
| 192 |
padding: 1em 1em 0; |
| 193 |
margin-bottom: 1.5em; |
| 194 |
} |
| 195 |
.alignCenter.art-text pre { |
| 196 |
padding: 0; |
| 197 |
} |
| 198 |
.alignCenter { |
| 199 |
margin: 1em 0; |
| 200 |
} |
| 201 |
.alignCenter > *:first-child { |
| 202 |
border: none; |
| 203 |
/* this isn't optimal, but it's an existence proof. PrinceXML doesn't |
| 204 |
support flexbox yet. |
| 205 |
*/ |
| 206 |
display: table; |
| 207 |
margin: 0 auto; |
| 208 |
} |
| 209 |
|
| 210 |
/* lists */ |
| 211 |
ol, ul { |
| 212 |
padding: 0; |
| 213 |
margin: 0 0 1em 2em; |
| 214 |
} |
| 215 |
ol ol, ul ul, ol ul, ul ol { |
| 216 |
margin-left: 1em; |
| 217 |
} |
| 218 |
li { |
| 219 |
margin: 0 0 0.25em 0; |
| 220 |
} |
| 221 |
.ulCompact li { |
| 222 |
margin: 0; |
| 223 |
} |
| 224 |
ul.empty, .ulEmpty { |
| 225 |
list-style-type: none; |
| 226 |
} |
| 227 |
ul.empty li, .ulEmpty li { |
| 228 |
margin-top: 0.5em; |
| 229 |
} |
| 230 |
ul.ulBare, li.ulBare { |
| 231 |
margin-left: 0em !important; |
| 232 |
} |
| 233 |
ul.compact, .ulCompact, |
| 234 |
ol.compact, .olCompact { |
| 235 |
line-height: 100%; |
| 236 |
margin: 0 0 0 2em; |
| 237 |
} |
| 238 |
|
| 239 |
/* definition lists */ |
| 240 |
dl { |
| 241 |
} |
| 242 |
dl > dt { |
| 243 |
float: left; |
| 244 |
margin-right: 1em; |
| 245 |
} |
| 246 |
/* |
| 247 |
dl.nohang > dt { |
| 248 |
float: none; |
| 249 |
} |
| 250 |
*/ |
| 251 |
dl > dd { |
| 252 |
margin-bottom: .8em; |
| 253 |
min-height: 1.3em; |
| 254 |
} |
| 255 |
dl.compact > dd, .dlCompact > dd { |
| 256 |
margin-bottom: 0em; |
| 257 |
} |
| 258 |
dl > dd > dl { |
| 259 |
margin-top: 0.5em; |
| 260 |
margin-bottom: 0em; |
| 261 |
} |
| 262 |
|
| 263 |
/* links */ |
| 264 |
a { |
| 265 |
text-decoration: none; |
| 266 |
} |
| 267 |
a[href] { |
| 268 |
color: #22e; /* Arlen: WCAG 2019 */ |
| 269 |
} |
| 270 |
a[href]:hover { |
| 271 |
background-color: #f2f2f2; |
| 272 |
} |
| 273 |
figcaption a[href], |
| 274 |
a[href].selfRef { |
| 275 |
color: #222; |
| 276 |
} |
| 277 |
/* XXX probably not this: |
| 278 |
a.selfRef:hover { |
| 279 |
background-color: transparent; |
| 280 |
cursor: default; |
| 281 |
} */ |
| 282 |
|
| 283 |
/* Figures */ |
| 284 |
tt, code, pre, code { |
| 285 |
background-color: #f9f9f9; |
| 286 |
font-family: 'Roboto Mono', monospace; |
| 287 |
} |
| 288 |
pre { |
| 289 |
border: 1px solid #eee; |
| 290 |
margin: 0; |
| 291 |
padding: 1em; |
| 292 |
} |
| 293 |
img { |
| 294 |
max-width: 100%; |
| 295 |
} |
| 296 |
figure { |
| 297 |
margin: 0; |
| 298 |
} |
| 299 |
figure blockquote { |
| 300 |
margin: 0.8em 0.4em 0.4em; |
| 301 |
} |
| 302 |
figcaption { |
| 303 |
font-style: italic; |
| 304 |
margin: 0 0 1em 0; |
| 305 |
} |
| 306 |
@media screen { |
| 307 |
pre { |
| 308 |
overflow-x: auto; |
| 309 |
max-width: 100%; |
| 310 |
max-width: calc(100% - 22px); |
| 311 |
} |
| 312 |
} |
| 313 |
|
| 314 |
/* aside, blockquote */ |
| 315 |
aside, blockquote { |
| 316 |
margin-left: 0; |
| 317 |
padding: 1.2em 2em; |
| 318 |
} |
| 319 |
blockquote { |
| 320 |
background-color: #f9f9f9; |
| 321 |
color: #111; /* Arlen: WCAG 2019 */ |
| 322 |
border: 1px solid #ddd; |
| 323 |
border-radius: 3px; |
| 324 |
margin: 1em 0; |
| 325 |
} |
| 326 |
cite { |
| 327 |
display: block; |
| 328 |
text-align: right; |
| 329 |
font-style: italic; |
| 330 |
} |
| 331 |
|
| 332 |
/* tables */ |
| 333 |
table { |
| 334 |
width: 100%; |
| 335 |
margin: 0 0 1em; |
| 336 |
border-collapse: collapse; |
| 337 |
border: 1px solid #eee; |
| 338 |
} |
| 339 |
th, td { |
| 340 |
text-align: left; |
| 341 |
vertical-align: top; |
| 342 |
padding: 0.5em 0.75em; |
| 343 |
} |
| 344 |
th { |
| 345 |
text-align: left; |
| 346 |
background-color: #e9e9e9; |
| 347 |
} |
| 348 |
tr:nth-child(2n+1) > td { |
| 349 |
background-color: #f5f5f5; |
| 350 |
} |
| 351 |
table caption { |
| 352 |
font-style: italic; |
| 353 |
margin: 0; |
| 354 |
padding: 0; |
| 355 |
text-align: left; |
| 356 |
} |
| 357 |
table p { |
| 358 |
/* XXX to avoid bottom margin on table row signifiers. If paragraphs should |
| 359 |
be allowed within tables more generally, it would be far better to select on a class. */ |
| 360 |
margin: 0; |
| 361 |
} |
| 362 |
|
| 363 |
/* pilcrow */ |
| 364 |
a.pilcrow { |
| 365 |
color: #666; /* Arlen: AHDJ 2019 */ |
| 366 |
text-decoration: none; |
| 367 |
visibility: hidden; |
| 368 |
user-select: none; |
| 369 |
-ms-user-select: none; |
| 370 |
-o-user-select:none; |
| 371 |
-moz-user-select: none; |
| 372 |
-khtml-user-select: none; |
| 373 |
-webkit-user-select: none; |
| 374 |
-webkit-touch-callout: none; |
| 375 |
} |
| 376 |
@media screen { |
| 377 |
aside:hover > a.pilcrow, |
| 378 |
p:hover > a.pilcrow, |
| 379 |
blockquote:hover > a.pilcrow, |
| 380 |
div:hover > a.pilcrow, |
| 381 |
li:hover > a.pilcrow, |
| 382 |
pre:hover > a.pilcrow { |
| 383 |
visibility: visible; |
| 384 |
} |
| 385 |
a.pilcrow:hover { |
| 386 |
background-color: transparent; |
| 387 |
} |
| 388 |
} |
| 389 |
|
| 390 |
/* misc */ |
| 391 |
hr { |
| 392 |
border: 0; |
| 393 |
border-top: 1px solid #eee; |
| 394 |
} |
| 395 |
.bcp14 { |
| 396 |
font-variant: small-caps; |
| 397 |
} |
| 398 |
|
| 399 |
.role { |
| 400 |
font-variant: all-small-caps; |
| 401 |
} |
| 402 |
|
| 403 |
/* info block */ |
| 404 |
#identifiers { |
| 405 |
margin: 0; |
| 406 |
font-size: 0.9em; |
| 407 |
} |
| 408 |
#identifiers dt { |
| 409 |
width: 3em; |
| 410 |
clear: left; |
| 411 |
} |
| 412 |
#identifiers dd { |
| 413 |
float: left; |
| 414 |
margin-bottom: 0; |
| 415 |
} |
| 416 |
#identifiers .authors .author { |
| 417 |
display: inline-block; |
| 418 |
margin-right: 1.5em; |
| 419 |
} |
| 420 |
#identifiers .authors .org { |
| 421 |
font-style: italic; |
| 422 |
} |
| 423 |
|
| 424 |
/* The prepared/rendered info at the very bottom of the page */ |
| 425 |
.docInfo { |
| 426 |
color: #666; /* Arlen: WCAG 2019 */ |
| 427 |
font-size: 0.9em; |
| 428 |
font-style: italic; |
| 429 |
margin-top: 2em; |
| 430 |
} |
| 431 |
.docInfo .prepared { |
| 432 |
float: left; |
| 433 |
} |
| 434 |
.docInfo .prepared { |
| 435 |
float: right; |
| 436 |
} |
| 437 |
|
| 438 |
/* table of contents */ |
| 439 |
#toc { |
| 440 |
padding: 0.75em 0 2em 0; |
| 441 |
margin-bottom: 1em; |
| 442 |
} |
| 443 |
nav.toc ul { |
| 444 |
margin: 0 0.5em 0 0; |
| 445 |
padding: 0; |
| 446 |
list-style: none; |
| 447 |
} |
| 448 |
nav.toc li { |
| 449 |
line-height: 1.3em; |
| 450 |
margin: 0.75em 0; |
| 451 |
padding-left: 1.2em; |
| 452 |
text-indent: -1.2em; |
| 453 |
} |
| 454 |
/* references */ |
| 455 |
.references dt { |
| 456 |
text-align: right; |
| 457 |
font-weight: bold; |
| 458 |
min-width: 7em; |
| 459 |
} |
| 460 |
.references dd { |
| 461 |
margin-left: 8em; |
| 462 |
overflow: auto; |
| 463 |
} |
| 464 |
|
| 465 |
.refInstance { |
| 466 |
margin-bottom: 1.25em; |
| 467 |
} |
| 468 |
|
| 469 |
.references .ascii { |
| 470 |
margin-bottom: 0.25em; |
| 471 |
} |
| 472 |
|
| 473 |
/* index */ |
| 474 |
.index ul { |
| 475 |
margin: 0 0 0 1em; |
| 476 |
padding: 0; |
| 477 |
list-style: none; |
| 478 |
} |
| 479 |
.index ul ul { |
| 480 |
margin: 0; |
| 481 |
} |
| 482 |
.index li { |
| 483 |
margin: 0; |
| 484 |
text-indent: -2em; |
| 485 |
padding-left: 2em; |
| 486 |
padding-bottom: 5px; |
| 487 |
} |
| 488 |
.indexIndex { |
| 489 |
margin: 0.5em 0 1em; |
| 490 |
} |
| 491 |
.index a { |
| 492 |
font-weight: 700; |
| 493 |
} |
| 494 |
/* make the index two-column on all but the smallest screens */ |
| 495 |
@media (min-width: 600px) { |
| 496 |
.index ul { |
| 497 |
-moz-column-count: 2; |
| 498 |
-moz-column-gap: 20px; |
| 499 |
} |
| 500 |
.index ul ul { |
| 501 |
-moz-column-count: 1; |
| 502 |
-moz-column-gap: 0; |
| 503 |
} |
| 504 |
} |
| 505 |
|
| 506 |
/* authors */ |
| 507 |
address.vcard { |
| 508 |
font-style: normal; |
| 509 |
margin: 1em 0; |
| 510 |
} |
| 511 |
|
| 512 |
address.vcard .nameRole { |
| 513 |
font-weight: 700; |
| 514 |
margin-left: 0; |
| 515 |
} |
| 516 |
address.vcard .label { |
| 517 |
font-family: "Noto Sans",Arial,Helvetica,sans-serif; |
| 518 |
margin: 0.5em 0; |
| 519 |
} |
| 520 |
address.vcard .type { |
| 521 |
display: none; |
| 522 |
} |
| 523 |
.alternative-contact { |
| 524 |
margin: 1.5em 0 1em; |
| 525 |
} |
| 526 |
hr.addr { |
| 527 |
border-top: 1px dashed; |
| 528 |
margin: 0; |
| 529 |
color: #ddd; |
| 530 |
max-width: calc(100% - 16px); |
| 531 |
} |
| 532 |
|
| 533 |
/* temporary notes */ |
| 534 |
.rfcEditorRemove::before { |
| 535 |
position: absolute; |
| 536 |
top: 0.2em; |
| 537 |
right: 0.2em; |
| 538 |
padding: 0.2em; |
| 539 |
content: "The RFC Editor will remove this note"; |
| 540 |
color: #9e2a00; /* Arlen: WCAG 2019 */ |
| 541 |
background-color: #ffd; /* Arlen: WCAG 2019 */ |
| 542 |
} |
| 543 |
.rfcEditorRemove { |
| 544 |
position: relative; |
| 545 |
padding-top: 1.8em; |
| 546 |
background-color: #ffd; /* Arlen: WCAG 2019 */ |
| 547 |
border-radius: 3px; |
| 548 |
} |
| 549 |
.cref { |
| 550 |
background-color: #ffd; /* Arlen: WCAG 2019 */ |
| 551 |
padding: 2px 4px; |
| 552 |
} |
| 553 |
.crefSource { |
| 554 |
font-style: italic; |
| 555 |
} |
| 556 |
/* alternative layout for smaller screens */ |
| 557 |
@media screen and (max-width: 1023px) { |
| 558 |
body { |
| 559 |
padding-top: 2em; |
| 560 |
} |
| 561 |
#title { |
| 562 |
padding: 1em 0; |
| 563 |
} |
| 564 |
h1 { |
| 565 |
font-size: 24px; |
| 566 |
} |
| 567 |
h2 { |
| 568 |
font-size: 20px; |
| 569 |
margin-top: -18px; /* provide offset for in-page anchors */ |
| 570 |
padding-top: 38px; |
| 571 |
} |
| 572 |
#identifiers dd { |
| 573 |
max-width: 60%; |
| 574 |
} |
| 575 |
#toc { |
| 576 |
position: fixed; |
| 577 |
z-index: 2; |
| 578 |
top: 0; |
| 579 |
right: 0; |
| 580 |
padding: 0; |
| 581 |
margin: 0; |
| 582 |
background-color: inherit; |
| 583 |
border-bottom: 1px solid #ccc; |
| 584 |
} |
| 585 |
#toc h2 { |
| 586 |
margin: -1px 0 0 0; |
| 587 |
padding: 4px 0 4px 6px; |
| 588 |
padding-right: 1em; |
| 589 |
min-width: 190px; |
| 590 |
font-size: 1.1em; |
| 591 |
text-align: right; |
| 592 |
background-color: #444; |
| 593 |
color: white; |
| 594 |
cursor: pointer; |
| 595 |
} |
| 596 |
#toc h2::before { /* css hamburger */ |
| 597 |
float: right; |
| 598 |
position: relative; |
| 599 |
width: 1em; |
| 600 |
height: 1px; |
| 601 |
left: -164px; |
| 602 |
margin: 6px 0 0 0; |
| 603 |
background: white none repeat scroll 0 0; |
| 604 |
box-shadow: 0 4px 0 0 white, 0 8px 0 0 white; |
| 605 |
content: ""; |
| 606 |
} |
| 607 |
#toc nav { |
| 608 |
display: none; |
| 609 |
padding: 0.5em 1em 1em; |
| 610 |
overflow: auto; |
| 611 |
height: calc(100vh - 48px); |
| 612 |
border-left: 1px solid #ddd; |
| 613 |
} |
| 614 |
} |
| 615 |
|
| 616 |
/* alternative layout for wide screens */ |
| 617 |
@media screen and (min-width: 1024px) { |
| 618 |
body { |
| 619 |
max-width: 724px; |
| 620 |
margin: 42px auto; |
| 621 |
padding-left: 1.5em; |
| 622 |
padding-right: 29em; |
| 623 |
} |
| 624 |
#toc { |
| 625 |
position: fixed; |
| 626 |
top: 42px; |
| 627 |
right: 42px; |
| 628 |
width: 25%; |
| 629 |
margin: 0; |
| 630 |
padding: 0 1em; |
| 631 |
z-index: 1; |
| 632 |
} |
| 633 |
#toc h2 { |
| 634 |
border-top: none; |
| 635 |
border-bottom: 1px solid #ddd; |
| 636 |
font-size: 1em; |
| 637 |
font-weight: normal; |
| 638 |
margin: 0; |
| 639 |
padding: 0.25em 1em 1em 0; |
| 640 |
} |
| 641 |
#toc nav { |
| 642 |
display: block; |
| 643 |
height: calc(90vh - 84px); |
| 644 |
bottom: 0; |
| 645 |
padding: 0.5em 0 0; |
| 646 |
overflow: auto; |
| 647 |
} |
| 648 |
img { /* future proofing */ |
| 649 |
max-width: 100%; |
| 650 |
height: auto; |
| 651 |
} |
| 652 |
} |
| 653 |
|
| 654 |
/* pagination */ |
| 655 |
@media print { |
| 656 |
body { |
| 657 |
|
| 658 |
width: 100%; |
| 659 |
} |
| 660 |
p { |
| 661 |
orphans: 3; |
| 662 |
widows: 3; |
| 663 |
} |
| 664 |
#n-copyright-notice { |
| 665 |
border-bottom: none; |
| 666 |
} |
| 667 |
#toc, #n-introduction { |
| 668 |
page-break-before: always; |
| 669 |
} |
| 670 |
#toc { |
| 671 |
border-top: none; |
| 672 |
padding-top: 0; |
| 673 |
} |
| 674 |
figure, pre { |
| 675 |
page-break-inside: avoid; |
| 676 |
} |
| 677 |
figure { |
| 678 |
overflow: scroll; |
| 679 |
} |
| 680 |
h1, h2, h3, h4, h5, h6 { |
| 681 |
page-break-after: avoid; |
| 682 |
} |
| 683 |
h2+*, h3+*, h4+*, h5+*, h6+* { |
| 684 |
page-break-before: avoid; |
| 685 |
} |
| 686 |
pre { |
| 687 |
white-space: pre-wrap; |
| 688 |
word-wrap: break-word; |
| 689 |
font-size: 10pt; |
| 690 |
} |
| 691 |
table { |
| 692 |
border: 1px solid #ddd; |
| 693 |
} |
| 694 |
td { |
| 695 |
border-top: 1px solid #ddd; |
| 696 |
} |
| 697 |
} |
| 698 |
|
| 699 |
/* This is commented out here, as the string-set: doesn't |
| 700 |
pass W3C validation currently */ |
| 701 |
/* |
| 702 |
.ears thead .left { |
| 703 |
string-set: ears-top-left content(); |
| 704 |
} |
| 705 |
|
| 706 |
.ears thead .center { |
| 707 |
string-set: ears-top-center content(); |
| 708 |
} |
| 709 |
|
| 710 |
.ears thead .right { |
| 711 |
string-set: ears-top-right content(); |
| 712 |
} |
| 713 |
|
| 714 |
.ears tfoot .left { |
| 715 |
string-set: ears-bottom-left content(); |
| 716 |
} |
| 717 |
|
| 718 |
.ears tfoot .center { |
| 719 |
string-set: ears-bottom-center content(); |
| 720 |
} |
| 721 |
|
| 722 |
.ears tfoot .right { |
| 723 |
string-set: ears-bottom-right content(); |
| 724 |
} |
| 725 |
*/ |
| 726 |
|
| 727 |
@page :first { |
| 728 |
padding-top: 0; |
| 729 |
@top-left { |
| 730 |
content: normal; |
| 731 |
border: none; |
| 732 |
} |
| 733 |
@top-center { |
| 734 |
content: normal; |
| 735 |
border: none; |
| 736 |
} |
| 737 |
@top-right { |
| 738 |
content: normal; |
| 739 |
border: none; |
| 740 |
} |
| 741 |
} |
| 742 |
|
| 743 |
@page { |
| 744 |
size: A4; |
| 745 |
margin-bottom: 45mm; |
| 746 |
padding-top: 20px; |
| 747 |
/* The follwing is commented out here, but set appropriately by in code, as |
| 748 |
the content depends on the document */ |
| 749 |
/* |
| 750 |
@top-left { |
| 751 |
content: 'Internet-Draft'; |
| 752 |
vertical-align: bottom; |
| 753 |
border-bottom: solid 1px #ccc; |
| 754 |
} |
| 755 |
@top-left { |
| 756 |
content: string(ears-top-left); |
| 757 |
vertical-align: bottom; |
| 758 |
border-bottom: solid 1px #ccc; |
| 759 |
} |
| 760 |
@top-center { |
| 761 |
content: string(ears-top-center); |
| 762 |
vertical-align: bottom; |
| 763 |
border-bottom: solid 1px #ccc; |
| 764 |
} |
| 765 |
@top-right { |
| 766 |
content: string(ears-top-right); |
| 767 |
vertical-align: bottom; |
| 768 |
border-bottom: solid 1px #ccc; |
| 769 |
} |
| 770 |
@bottom-left { |
| 771 |
content: string(ears-bottom-left); |
| 772 |
vertical-align: top; |
| 773 |
border-top: solid 1px #ccc; |
| 774 |
} |
| 775 |
@bottom-center { |
| 776 |
content: string(ears-bottom-center); |
| 777 |
vertical-align: top; |
| 778 |
border-top: solid 1px #ccc; |
| 779 |
} |
| 780 |
@bottom-right { |
| 781 |
content: '[Page ' counter(page) ']'; |
| 782 |
vertical-align: top; |
| 783 |
border-top: solid 1px #ccc; |
| 784 |
} |
| 785 |
*/ |
| 786 |
|
| 787 |
} |
| 788 |
|
| 789 |
/* Changes introduced to fix issues found during implementation */ |
| 790 |
/* Make sure links are clickable even if overlapped by following H* */ |
| 791 |
a { |
| 792 |
z-index: 2; |
| 793 |
} |
| 794 |
/* Separate body from document info even without intervening H1 */ |
| 795 |
section { |
| 796 |
clear: both; |
| 797 |
} |
| 798 |
|
| 799 |
|
| 800 |
/* Top align author divs, to avoid names without organization dropping level with org names */ |
| 801 |
.author { |
| 802 |
vertical-align: top; |
| 803 |
} |
| 804 |
|
| 805 |
/* Leave room in document info to show Internet-Draft on one line */ |
| 806 |
#identifiers dt { |
| 807 |
width: 8em; |
| 808 |
} |
| 809 |
|
| 810 |
/* Don't waste quite as much whitespace between label and value in doc info */ |
| 811 |
#identifiers dd { |
| 812 |
margin-left: 1em; |
| 813 |
} |
| 814 |
|
| 815 |
/* Give floating toc a background color (needed when it's a div inside section */ |
| 816 |
#toc { |
| 817 |
background-color: white; |
| 818 |
} |
| 819 |
|
| 820 |
/* Make the collapsed ToC header render white on gray also when it's a link */ |
| 821 |
@media screen and (max-width: 1023px) { |
| 822 |
#toc h2 a, |
| 823 |
#toc h2 a:link, |
| 824 |
#toc h2 a:focus, |
| 825 |
#toc h2 a:hover, |
| 826 |
#toc a.toplink, |
| 827 |
#toc a.toplink:hover { |
| 828 |
color: white; |
| 829 |
background-color: #444; |
| 830 |
text-decoration: none; |
| 831 |
} |
| 832 |
} |
| 833 |
|
| 834 |
/* Give the bottom of the ToC some whitespace */ |
| 835 |
@media screen and (min-width: 1024px) { |
| 836 |
#toc { |
| 837 |
padding: 0 0 1em 1em; |
| 838 |
} |
| 839 |
} |
| 840 |
|
| 841 |
/* Style section numbers with more space between number and title */ |
| 842 |
.section-number { |
| 843 |
padding-right: 0.5em; |
| 844 |
} |
| 845 |
|
| 846 |
/* prevent monospace from becoming overly large */ |
| 847 |
tt, code, pre, code { |
| 848 |
font-size: 95%; |
| 849 |
} |
| 850 |
|
| 851 |
/* Fix the height/width aspect for ascii art*/ |
| 852 |
pre.sourcecode, |
| 853 |
.art-text pre { |
| 854 |
line-height: 1.12; |
| 855 |
} |
| 856 |
|
| 857 |
|
| 858 |
/* Add styling for a link in the ToC that points to the top of the document */ |
| 859 |
a.toplink { |
| 860 |
float: right; |
| 861 |
margin-right: 0.5em; |
| 862 |
} |
| 863 |
|
| 864 |
/* Fix the dl styling to match the RFC 7992 attributes */ |
| 865 |
dl > dt, |
| 866 |
dl.dlParallel > dt { |
| 867 |
float: left; |
| 868 |
margin-right: 1em; |
| 869 |
} |
| 870 |
dl.dlNewline > dt { |
| 871 |
float: none; |
| 872 |
} |
| 873 |
|
| 874 |
/* Provide styling for table cell text alignment */ |
| 875 |
table td.text-left, |
| 876 |
table th.text-left { |
| 877 |
text-align: left; |
| 878 |
} |
| 879 |
table td.text-center, |
| 880 |
table th.text-center { |
| 881 |
text-align: center; |
| 882 |
} |
| 883 |
table td.text-right, |
| 884 |
table th.text-right { |
| 885 |
text-align: right; |
| 886 |
} |
| 887 |
|
| 888 |
/* Make the alternative author contact informatio look less like just another |
| 889 |
author, and group it closer with the primary author contact information */ |
| 890 |
.alternative-contact { |
| 891 |
margin: 0.5em 0 0.25em 0; |
| 892 |
} |
| 893 |
address .non-ascii { |
| 894 |
margin: 0 0 0 2em; |
| 895 |
} |
| 896 |
|
| 897 |
/* With it being possible to set tables with alignment |
| 898 |
left, center, and right, { width: 100%; } does not make sense */ |
| 899 |
table { |
| 900 |
width: auto; |
| 901 |
} |
| 902 |
|
| 903 |
/* Avoid reference text that sits in a block with very wide left margin, |
| 904 |
because of a long floating dt label.*/ |
| 905 |
.references dd { |
| 906 |
overflow: visible; |
| 907 |
} |
| 908 |
|
| 909 |
/* Control caption placement */ |
| 910 |
caption { |
| 911 |
caption-side: bottom; |
| 912 |
} |
| 913 |
|
| 914 |
/* Limit the width of the author address vcard, so names in right-to-left |
| 915 |
script don't end up on the other side of the page. */ |
| 916 |
|
| 917 |
address.vcard { |
| 918 |
max-width: 30em; |
| 919 |
margin-right: auto; |
| 920 |
} |
| 921 |
|
| 922 |
/* For address alignment dependent on LTR or RTL scripts */ |
| 923 |
address div.left { |
| 924 |
text-align: left; |
| 925 |
} |
| 926 |
address div.right { |
| 927 |
text-align: right; |
| 928 |
} |
| 929 |
|
| 930 |
/* Provide table alignment support. We can't use the alignX classes above |
| 931 |
since they do unwanted things with caption and other styling. */ |
| 932 |
table.right { |
| 933 |
margin-left: auto; |
| 934 |
margin-right: 0; |
| 935 |
} |
| 936 |
table.center { |
| 937 |
margin-left: auto; |
| 938 |
margin-right: auto; |
| 939 |
} |
| 940 |
table.left { |
| 941 |
margin-left: 0; |
| 942 |
margin-right: auto; |
| 943 |
} |
| 944 |
|
| 945 |
/* Give the table caption label the same styling as the figcaption */ |
| 946 |
caption a[href] { |
| 947 |
color: #222; |
| 948 |
} |
| 949 |
|
| 950 |
@media print { |
| 951 |
.toplink { |
| 952 |
display: none; |
| 953 |
} |
| 954 |
|
| 955 |
/* avoid overwriting the top border line with the ToC header */ |
| 956 |
#toc { |
| 957 |
padding-top: 1px; |
| 958 |
} |
| 959 |
|
| 960 |
/* Avoid page breaks inside dl and author address entries */ |
| 961 |
.vcard { |
| 962 |
page-break-inside: avoid; |
| 963 |
} |
| 964 |
|
| 965 |
} |
| 966 |
/* Tweak the bcp14 keyword presentation */ |
| 967 |
.bcp14 { |
| 968 |
font-variant: small-caps; |
| 969 |
font-weight: bold; |
| 970 |
font-size: 0.9em; |
| 971 |
} |
| 972 |
/* Tweak the invisible space above H* in order not to overlay links in text above */ |
| 973 |
h2 { |
| 974 |
margin-top: -18px; /* provide offset for in-page anchors */ |
| 975 |
padding-top: 31px; |
| 976 |
} |
| 977 |
h3 { |
| 978 |
margin-top: -18px; /* provide offset for in-page anchors */ |
| 979 |
padding-top: 24px; |
| 980 |
} |
| 981 |
h4 { |
| 982 |
margin-top: -18px; /* provide offset for in-page anchors */ |
| 983 |
padding-top: 24px; |
| 984 |
} |
| 985 |
/* Float artwork pilcrow to the right */ |
| 986 |
@media screen { |
| 987 |
.artwork a.pilcrow { |
| 988 |
display: block; |
| 989 |
line-height: 0.7; |
| 990 |
margin-top: 0.15em; |
| 991 |
} |
| 992 |
} |
| 993 |
/* Make pilcrows on dd visible */ |
| 994 |
@media screen { |
| 995 |
dd:hover > a.pilcrow { |
| 996 |
visibility: visible; |
| 997 |
} |
| 998 |
} |
| 999 |
/* Make the placement of figcaption match that of a table's caption |
| 1000 |
by removing the figure's added bottom margin */ |
| 1001 |
.alignLeft.art-text, |
| 1002 |
.alignCenter.art-text, |
| 1003 |
.alignRight.art-text { |
| 1004 |
margin-bottom: 0; |
| 1005 |
} |
| 1006 |
.alignLeft, |
| 1007 |
.alignCenter, |
| 1008 |
.alignRight { |
| 1009 |
margin: 1em 0 0 0; |
| 1010 |
} |
| 1011 |
/* In print, the pilcrow won't show on hover, so prevent it from taking up space, |
| 1012 |
possibly even requiring a new line */ |
| 1013 |
@media print { |
| 1014 |
a.pilcrow { |
| 1015 |
display: none; |
| 1016 |
} |
| 1017 |
} |
| 1018 |
/* Styling for the external metadata */ |
| 1019 |
div#external-metadata { |
| 1020 |
background-color: #eee; |
| 1021 |
padding: 0.5em; |
| 1022 |
margin-bottom: 0.5em; |
| 1023 |
display: none; |
| 1024 |
} |
| 1025 |
div#internal-metadata { |
| 1026 |
padding: 0.5em; /* to match the external-metadata padding */ |
| 1027 |
} |
| 1028 |
/* Styling for title RFC Number */ |
| 1029 |
h1#rfcnum { |
| 1030 |
clear: both; |
| 1031 |
margin: 0 0 -1em; |
| 1032 |
padding: 1em 0 0 0; |
| 1033 |
} |
| 1034 |
/* Make .olPercent look the same as <ol><li> */ |
| 1035 |
dl.olPercent > dd { |
| 1036 |
margin-bottom: 0.25em; |
| 1037 |
min-height: initial; |
| 1038 |
} |
| 1039 |
/* Give aside some styling to set it apart */ |
| 1040 |
aside { |
| 1041 |
border-left: 1px solid #ddd; |
| 1042 |
margin: 1em 0 1em 2em; |
| 1043 |
padding: 0.2em 2em; |
| 1044 |
} |
| 1045 |
aside > dl, |
| 1046 |
aside > ol, |
| 1047 |
aside > ul, |
| 1048 |
aside > table, |
| 1049 |
aside > p { |
| 1050 |
margin-bottom: 0.5em; |
| 1051 |
} |
| 1052 |
/* Additional page break settings */ |
| 1053 |
@media print { |
| 1054 |
figcaption, table caption { |
| 1055 |
page-break-before: avoid; |
| 1056 |
} |
| 1057 |
} |
| 1058 |
/* Font size adjustments for print */ |
| 1059 |
@media print { |
| 1060 |
body { font-size: 10pt; line-height: normal; max-width: 96%; } |
| 1061 |
h1 { font-size: 1.72em; padding-top: 1.5em; } /* 1*1.2*1.2*1.2 */ |
| 1062 |
h2 { font-size: 1.44em; padding-top: 1.5em; } /* 1*1.2*1.2 */ |
| 1063 |
h3 { font-size: 1.2em; padding-top: 1.5em; } /* 1*1.2 */ |
| 1064 |
h4 { font-size: 1em; padding-top: 1.5em; } |
| 1065 |
h5, h6 { font-size: 1em; margin: initial; padding: 0.5em 0 0.3em; } |
| 1066 |
} |
| 1067 |
/* Sourcecode margin in print, when there's no pilcrow */ |
| 1068 |
@media print { |
| 1069 |
.artwork, |
| 1070 |
.sourcecode { |
| 1071 |
margin-bottom: 1em; |
| 1072 |
} |
| 1073 |
} |
| 1074 |
/* Avoid narrow tables forcing too narrow table captions, which may render badly */ |
| 1075 |
table { |
| 1076 |
min-width: 20em; |
| 1077 |
} |
| 1078 |
/* ol type a */ |
| 1079 |
ol.type-a { list-style-type: lower-alpha; } |
| 1080 |
ol.type-A { list-style-type: upper-alpha; } |
| 1081 |
ol.type-i { list-style-type: lower-roman; } |
| 1082 |
ol.type-I { list-style-type: lower-roman; } |
| 1083 |
/* Apply the print table and row borders in general, on request from the RPC, |
| 1084 |
and increase the contrast between border and odd row background sligthtly */ |
| 1085 |
table { |
| 1086 |
border: 1px solid #ddd; |
| 1087 |
} |
| 1088 |
td { |
| 1089 |
border-top: 1px solid #ddd; |
| 1090 |
} |
| 1091 |
tr:nth-child(2n+1) > td { |
| 1092 |
background-color: #f8f8f8; |
| 1093 |
} |
| 1094 |
/* Use style rules to govern display of the TOC. */ |
| 1095 |
@media screen and (max-width: 1023px) { |
| 1096 |
#toc nav { display: none; } |
| 1097 |
#toc.active nav { display: block; } |
| 1098 |
} |
| 1099 |
/* Add support for keepWithNext */ |
| 1100 |
.keepWithNext { |
| 1101 |
break-after: avoid-page; |
| 1102 |
break-after: avoid-page; |
| 1103 |
} |
| 1104 |
/* Add support for keepWithPrevious */ |
| 1105 |
.keepWithPrevious { |
| 1106 |
break-before: avoid-page; |
| 1107 |
} |
| 1108 |
/* Change the approach to avoiding breaks inside artwork etc. */ |
| 1109 |
figure, pre, table, .artwork, .sourcecode { |
| 1110 |
break-before: avoid-page; |
| 1111 |
break-after: auto; |
| 1112 |
} |
| 1113 |
/* Avoid breaks between <dt> and <dd> */ |
| 1114 |
dl { |
| 1115 |
break-before: auto; |
| 1116 |
break-inside: auto; |
| 1117 |
} |
| 1118 |
dt { |
| 1119 |
break-before: auto; |
| 1120 |
break-after: avoid-page; |
| 1121 |
} |
| 1122 |
dd { |
| 1123 |
break-before: avoid-page; |
| 1124 |
break-after: auto; |
| 1125 |
orphans: 3; |
| 1126 |
widows: 3 |
| 1127 |
} |
| 1128 |
span.break, dd.break { |
| 1129 |
margin-bottom: 0; |
| 1130 |
min-height: 0; |
| 1131 |
break-before: auto; |
| 1132 |
break-inside: auto; |
| 1133 |
break-after: auto; |
| 1134 |
} |
| 1135 |
/* Undo break-before ToC */ |
| 1136 |
@media print { |
| 1137 |
#toc { |
| 1138 |
break-before: auto; |
| 1139 |
} |
| 1140 |
} |
| 1141 |
/* Text in compact lists should not get extra bottim margin space, |
| 1142 |
since that would makes the list not compact */ |
| 1143 |
ul.compact p, .ulCompact p, |
| 1144 |
ol.compact p, .olCompact p { |
| 1145 |
margin: 0; |
| 1146 |
} |
| 1147 |
/* But the list as a whole needs the extra space at the end */ |
| 1148 |
section ul.compact, |
| 1149 |
section .ulCompact, |
| 1150 |
section ol.compact, |
| 1151 |
section .olCompact { |
| 1152 |
margin-bottom: 1em; /* same as p not within ul.compact etc. */ |
| 1153 |
} |
| 1154 |
/* The tt and code background above interferes with for instance table cell |
| 1155 |
backgrounds. Changed to something a bit more selective. */ |
| 1156 |
tt, code { |
| 1157 |
background-color: transparent; |
| 1158 |
} |
| 1159 |
p tt, p code, li tt, li code { |
| 1160 |
background-color: #f8f8f8; |
| 1161 |
} |
| 1162 |
/* Tweak the pre margin -- 0px doesn't come out well */ |
| 1163 |
pre { |
| 1164 |
margin-top: 0.5px; |
| 1165 |
} |
| 1166 |
/* Tweak the comact list text */ |
| 1167 |
ul.compact, .ulCompact, |
| 1168 |
ol.compact, .olCompact, |
| 1169 |
dl.compact, .dlCompact { |
| 1170 |
line-height: normal; |
| 1171 |
} |
| 1172 |
/* Don't add top margin for nested lists */ |
| 1173 |
li > ul, li > ol, li > dl, |
| 1174 |
dd > ul, dd > ol, dd > dl, |
| 1175 |
dl > dd > dl { |
| 1176 |
margin-top: initial; |
| 1177 |
} |
| 1178 |
/* Elements that should not be rendered on the same line as a <dt> */ |
| 1179 |
/* This should match the element list in writer.text.TextWriter.render_dl() */ |
| 1180 |
dd > div.artwork:first-child, |
| 1181 |
dd > aside:first-child, |
| 1182 |
dd > figure:first-child, |
| 1183 |
dd > ol:first-child, |
| 1184 |
dd > div:first-child > pre.sourcecode, |
| 1185 |
dd > table:first-child, |
| 1186 |
dd > ul:first-child { |
| 1187 |
clear: left; |
| 1188 |
} |
| 1189 |
/* fix for weird browser behaviour when <dd/> is empty */ |
| 1190 |
dt+dd:empty::before{ |
| 1191 |
content: "\00a0"; |
| 1192 |
} |
| 1193 |
/* Make paragraph spacing inside <li> smaller than in body text, to fit better within the list */ |
| 1194 |
li > p { |
| 1195 |
margin-bottom: 0.5em |
| 1196 |
} |
| 1197 |
/* Don't let p margin spill out from inside list items */ |
| 1198 |
li > p:last-of-type { |
| 1199 |
margin-bottom: 0; |
| 1200 |
} |
| 1201 |
</style> |
| 1202 |
<link href="rfc-local.css" rel="stylesheet" type="text/css"> |
| 1203 |
<link href="https://dx.doi.org/10.17487/rfc9117" rel="alternate"> |
| 1204 |
<link href="urn:issn:2070-1721" rel="alternate"> |
| 1205 |
<link href="https://datatracker.ietf.org/doc/draft-ietf-idr-bgp-flowspec-oid-15" rel="prev"> |
| 1206 |
</head> |
| 1207 |
<body> |
| 1208 |
<script src="https://www.rfc-editor.org/js/metadata.min.js"></script> |
| 1209 |
<table class="ears"> |
| 1210 |
<thead><tr> |
| 1211 |
<td class="left">RFC 9117</td> |
| 1212 |
<td class="center">Revised Flowspec Validation Procedure</td> |
| 1213 |
<td class="right">August 2021</td> |
| 1214 |
</tr></thead> |
| 1215 |
<tfoot><tr> |
| 1216 |
<td class="left">Uttaro, et al.</td> |
| 1217 |
<td class="center">Standards Track</td> |
| 1218 |
<td class="right">[Page]</td> |
| 1219 |
</tr></tfoot> |
| 1220 |
</table> |
| 1221 |
<div id="external-metadata" class="document-information"></div> |
| 1222 |
<div id="internal-metadata" class="document-information"> |
| 1223 |
<dl id="identifiers"> |
| 1224 |
<dt class="label-stream">Stream:</dt> |
| 1225 |
<dd class="stream">Internet Engineering Task Force (IETF)</dd> |
| 1226 |
<dt class="label-rfc">RFC:</dt> |
| 1227 |
<dd class="rfc"><a href="https://www.rfc-editor.org/rfc/rfc9117" class="eref">9117</a></dd> |
| 1228 |
<dt class="label-updates">Updates:</dt> |
| 1229 |
<dd class="updates"> |
| 1230 |
<a href="https://www.rfc-editor.org/rfc/rfc8955" class="eref">8955</a> </dd> |
| 1231 |
<dt class="label-category">Category:</dt> |
| 1232 |
<dd class="category">Standards Track</dd> |
| 1233 |
<dt class="label-published">Published:</dt> |
| 1234 |
<dd class="published"> |
| 1235 |
<time datetime="2021-08" class="published">August 2021</time> |
| 1236 |
</dd> |
| 1237 |
<dt class="label-issn">ISSN:</dt> |
| 1238 |
<dd class="issn">2070-1721</dd> |
| 1239 |
<dt class="label-authors">Authors:</dt> |
| 1240 |
<dd class="authors"> |
| 1241 |
<div class="author"> |
| 1242 |
<div class="author-name">J. Uttaro</div> |
| 1243 |
<div class="org">AT&T</div> |
| 1244 |
</div> |
| 1245 |
<div class="author"> |
| 1246 |
<div class="author-name">J. Alcaide</div> |
| 1247 |
<div class="org">Cisco</div> |
| 1248 |
</div> |
| 1249 |
<div class="author"> |
| 1250 |
<div class="author-name">C. Filsfils</div> |
| 1251 |
<div class="org">Cisco</div> |
| 1252 |
</div> |
| 1253 |
<div class="author"> |
| 1254 |
<div class="author-name">D. Smith</div> |
| 1255 |
<div class="org">Cisco</div> |
| 1256 |
</div> |
| 1257 |
<div class="author"> |
| 1258 |
<div class="author-name">P. Mohapatra</div> |
| 1259 |
<div class="org">Sproute Networks</div> |
| 1260 |
</div> |
| 1261 |
</dd> |
| 1262 |
</dl> |
| 1263 |
</div> |
| 1264 |
<h1 id="rfcnum">RFC 9117</h1> |
| 1265 |
<h1 id="title">Revised Validation Procedure for BGP Flow Specifications</h1> |
| 1266 |
<section id="section-abstract"> |
| 1267 |
<h2 id="abstract"><a href="#abstract" class="selfRef">Abstract</a></h2> |
| 1268 |
<p id="section-abstract-1"> |
| 1269 |
This document describes a modification to the validation procedure defined |
| 1270 |
for the dissemination of BGP Flow Specifications. The dissemination of BGP |
| 1271 |
Flow Specifications as specified in RFC 8955 requires that the originator |
| 1272 |
of the Flow Specification match the originator of the best-match unicast |
| 1273 |
route for the destination prefix embedded in the Flow Specification. For an |
| 1274 |
Internal Border Gateway Protocol (iBGP) received route, the originator is |
| 1275 |
typically a border router within the same autonomous system (AS). The |
| 1276 |
objective is to allow only BGP speakers within the data forwarding path to |
| 1277 |
originate BGP Flow Specifications. Sometimes it is desirable to originate |
| 1278 |
the BGP Flow Specification from any place within the autonomous system |
| 1279 |
itself, for example, from a centralized BGP route controller. However, the |
| 1280 |
validation procedure described in RFC 8955 will fail in this scenario. The modification |
| 1281 |
proposed herein relaxes the validation rule to enable Flow Specifications |
| 1282 |
to be originated within the same autonomous system as the BGP speaker |
| 1283 |
performing the validation. Additionally, this document revises the AS_PATH |
| 1284 |
validation rules so Flow Specifications received from an External Border |
| 1285 |
Gateway Protocol (eBGP) peer can be validated when such a peer is a BGP |
| 1286 |
route server.<a href="#section-abstract-1" class="pilcrow">¶</a></p> |
| 1287 |
<p id="section-abstract-2"> |
| 1288 |
This document updates the validation procedure in RFC 8955.<a href="#section-abstract-2" class="pilcrow">¶</a></p> |
| 1289 |
</section> |
| 1290 |
<div id="status-of-memo"> |
| 1291 |
<section id="section-boilerplate.1"> |
| 1292 |
<h2 id="name-status-of-this-memo"> |
| 1293 |
<a href="#name-status-of-this-memo" class="section-name selfRef">Status of This Memo</a> |
| 1294 |
</h2> |
| 1295 |
<p id="section-boilerplate.1-1"> |
| 1296 |
This is an Internet Standards Track document.<a href="#section-boilerplate.1-1" class="pilcrow">¶</a></p> |
| 1297 |
<p id="section-boilerplate.1-2"> |
| 1298 |
This document is a product of the Internet Engineering Task Force |
| 1299 |
(IETF). It represents the consensus of the IETF community. It has |
| 1300 |
received public review and has been approved for publication by |
| 1301 |
the Internet Engineering Steering Group (IESG). Further |
| 1302 |
information on Internet Standards is available in Section 2 of |
| 1303 |
RFC 7841.<a href="#section-boilerplate.1-2" class="pilcrow">¶</a></p> |
| 1304 |
<p id="section-boilerplate.1-3"> |
| 1305 |
Information about the current status of this document, any |
| 1306 |
errata, and how to provide feedback on it may be obtained at |
| 1307 |
<span><a href="https://www.rfc-editor.org/info/rfc9117">https://www.rfc-editor.org/info/rfc9117</a></span>.<a href="#section-boilerplate.1-3" class="pilcrow">¶</a></p> |
| 1308 |
</section> |
| 1309 |
</div> |
| 1310 |
<div id="copyright"> |
| 1311 |
<section id="section-boilerplate.2"> |
| 1312 |
<h2 id="name-copyright-notice"> |
| 1313 |
<a href="#name-copyright-notice" class="section-name selfRef">Copyright Notice</a> |
| 1314 |
</h2> |
| 1315 |
<p id="section-boilerplate.2-1"> |
| 1316 |
Copyright (c) 2021 IETF Trust and the persons identified as the |
| 1317 |
document authors. All rights reserved.<a href="#section-boilerplate.2-1" class="pilcrow">¶</a></p> |
| 1318 |
<p id="section-boilerplate.2-2"> |
| 1319 |
This document is subject to BCP 78 and the IETF Trust's Legal |
| 1320 |
Provisions Relating to IETF Documents |
| 1321 |
(<span><a href="https://trustee.ietf.org/license-info">https://trustee.ietf.org/license-info</a></span>) in effect on the date of |
| 1322 |
publication of this document. Please review these documents |
| 1323 |
carefully, as they describe your rights and restrictions with |
| 1324 |
respect to this document. Code Components extracted from this |
| 1325 |
document must include Simplified BSD License text as described in |
| 1326 |
Section 4.e of the Trust Legal Provisions and are provided without |
| 1327 |
warranty as described in the Simplified BSD License.<a href="#section-boilerplate.2-2" class="pilcrow">¶</a></p> |
| 1328 |
</section> |
| 1329 |
</div> |
| 1330 |
<div id="toc"> |
| 1331 |
<section id="section-toc.1"> |
| 1332 |
<a href="#" onclick="scroll(0,0)" class="toplink">▲</a><h2 id="name-table-of-contents"> |
| 1333 |
<a href="#name-table-of-contents" class="section-name selfRef">Table of Contents</a> |
| 1334 |
</h2> |
| 1335 |
<nav class="toc"><ul class="compact ulBare ulEmpty toc"> |
| 1336 |
<li class="compact ulBare ulEmpty toc" id="section-toc.1-1.1"> |
| 1337 |
<p id="section-toc.1-1.1.1" class="keepWithNext"><a href="#section-1" class="xref">1</a>. <a href="#name-introduction" class="xref">Introduction</a></p> |
| 1338 |
</li> |
| 1339 |
<li class="compact ulBare ulEmpty toc" id="section-toc.1-1.2"> |
| 1340 |
<p id="section-toc.1-1.2.1" class="keepWithNext"><a href="#section-2" class="xref">2</a>. <a href="#name-definitions-of-terms-used-i" class="xref">Definitions of Terms Used in This Memo</a></p> |
| 1341 |
</li> |
| 1342 |
<li class="compact ulBare ulEmpty toc" id="section-toc.1-1.3"> |
| 1343 |
<p id="section-toc.1-1.3.1" class="keepWithNext"><a href="#section-3" class="xref">3</a>. <a href="#name-motivation" class="xref">Motivation</a></p> |
| 1344 |
</li> |
| 1345 |
<li class="compact ulBare ulEmpty toc" id="section-toc.1-1.4"> |
| 1346 |
<p id="section-toc.1-1.4.1"><a href="#section-4" class="xref">4</a>. <a href="#name-revised-validation-procedur" class="xref">Revised Validation Procedure</a></p> |
| 1347 |
<ul class="ulEmpty compact ulBare toc"> |
| 1348 |
<li class="ulEmpty compact ulBare toc" id="section-toc.1-1.4.2.1"> |
| 1349 |
<p id="section-toc.1-1.4.2.1.1"><a href="#section-4.1" class="xref">4.1</a>. <a href="#name-revision-of-route-feasibili" class="xref">Revision of Route Feasibility</a></p> |
| 1350 |
</li> |
| 1351 |
<li class="ulEmpty compact ulBare toc" id="section-toc.1-1.4.2.2"> |
| 1352 |
<p id="section-toc.1-1.4.2.2.1"><a href="#section-4.2" class="xref">4.2</a>. <a href="#name-revision-of-as_path-validat" class="xref">Revision of AS_PATH Validation</a></p> |
| 1353 |
</li> |
| 1354 |
</ul> |
| 1355 |
</li> |
| 1356 |
<li class="compact ulBare ulEmpty toc" id="section-toc.1-1.5"> |
| 1357 |
<p id="section-toc.1-1.5.1"><a href="#section-5" class="xref">5</a>. <a href="#name-topology-considerations" class="xref">Topology Considerations</a></p> |
| 1358 |
</li> |
| 1359 |
<li class="compact ulBare ulEmpty toc" id="section-toc.1-1.6"> |
| 1360 |
<p id="section-toc.1-1.6.1"><a href="#section-6" class="xref">6</a>. <a href="#name-iana-considerations" class="xref">IANA Considerations</a></p> |
| 1361 |
</li> |
| 1362 |
<li class="compact ulBare ulEmpty toc" id="section-toc.1-1.7"> |
| 1363 |
<p id="section-toc.1-1.7.1"><a href="#section-7" class="xref">7</a>. <a href="#name-security-considerations" class="xref">Security Considerations</a></p> |
| 1364 |
</li> |
| 1365 |
<li class="compact ulBare ulEmpty toc" id="section-toc.1-1.8"> |
| 1366 |
<p id="section-toc.1-1.8.1"><a href="#section-8" class="xref">8</a>. <a href="#name-references" class="xref">References</a></p> |
| 1367 |
<ul class="ulEmpty compact ulBare toc"> |
| 1368 |
<li class="ulEmpty compact ulBare toc" id="section-toc.1-1.8.2.1"> |
| 1369 |
<p id="section-toc.1-1.8.2.1.1"><a href="#section-8.1" class="xref">8.1</a>. <a href="#name-normative-references" class="xref">Normative References</a></p> |
| 1370 |
</li> |
| 1371 |
<li class="ulEmpty compact ulBare toc" id="section-toc.1-1.8.2.2"> |
| 1372 |
<p id="section-toc.1-1.8.2.2.1"><a href="#section-8.2" class="xref">8.2</a>. <a href="#name-informative-references" class="xref">Informative References</a></p> |
| 1373 |
</li> |
| 1374 |
</ul> |
| 1375 |
</li> |
| 1376 |
<li class="compact ulBare ulEmpty toc" id="section-toc.1-1.9"> |
| 1377 |
<p id="section-toc.1-1.9.1"><a href="#appendix-A" class="xref"></a><a href="#name-acknowledgements" class="xref">Acknowledgements</a></p> |
| 1378 |
</li> |
| 1379 |
<li class="compact ulBare ulEmpty toc" id="section-toc.1-1.10"> |
| 1380 |
<p id="section-toc.1-1.10.1"><a href="#appendix-B" class="xref"></a><a href="#name-authors-addresses" class="xref">Authors' Addresses</a></p> |
| 1381 |
</li> |
| 1382 |
</ul> |
| 1383 |
</nav> |
| 1384 |
</section> |
| 1385 |
</div> |
| 1386 |
<section id="section-1"> |
| 1387 |
<h2 id="name-introduction"> |
| 1388 |
<a href="#section-1" class="section-number selfRef">1. </a><a href="#name-introduction" class="section-name selfRef">Introduction</a> |
| 1389 |
</h2> |
| 1390 |
<p id="section-1-1"> |
| 1391 |
<span>[<a href="#RFC8955" class="xref">RFC8955</a>]</span> defines BGP Network Layer |
| 1392 |
Reachability Information (NLRI) <span>[<a href="#RFC4760" class="xref">RFC4760</a>]</span> that can be used to distribute traffic Flow |
| 1393 |
Specifications amongst BGP speakers in support of traffic |
| 1394 |
filtering. The primary intention of <span>[<a href="#RFC8955" class="xref">RFC8955</a>]</span> is to enable downstream autonomous systems to |
| 1395 |
signal traffic filtering policies to upstream autonomous systems. In |
| 1396 |
this way, traffic is filtered closer to the source and the upstream |
| 1397 |
autonomous systems avoid carrying the traffic to the downstream |
| 1398 |
autonomous systems only to be discarded. <span>[<a href="#RFC8955" class="xref">RFC8955</a>]</span> also enables more granular traffic filtering based |
| 1399 |
upon upper-layer protocol information (e.g., protocol or port |
| 1400 |
numbers) as opposed to coarse IP destination prefix-based filtering. |
| 1401 |
Flow Specification NLRIs received from a BGP peer is subject to |
| 1402 |
validity checks before being considered feasible and subsequently |
| 1403 |
installed within the respective Adj-RIB-In.<a href="#section-1-1" class="pilcrow">¶</a></p> |
| 1404 |
<p id="section-1-2"> |
| 1405 |
The validation procedure defined within <span>[<a href="#RFC8955" class="xref">RFC8955</a>]</span> requires that the originator of the Flow |
| 1406 |
Specification NLRI match the originator of the best-match unicast |
| 1407 |
route for the destination prefix embedded in the Flow Specification. |
| 1408 |
The aim is to make sure that only speakers on the forwarding path |
| 1409 |
can originate the Flow Specification. Let's consider the particular |
| 1410 |
case where the Flow Specification is originated in any location |
| 1411 |
within the same Local Domain as the speaker performing the |
| 1412 |
validation (for example, by a centralized BGP route controller), and |
| 1413 |
the best-match unicast route is originated in another Local Domain. |
| 1414 |
In order for the validation to succeed for a Flow Specification |
| 1415 |
received from an iBGP peer, it would be necessary to disseminate |
| 1416 |
such Flow Specification NLRI directly from the specific border |
| 1417 |
router (within the Local Domain) that is advertising the |
| 1418 |
corresponding best-match unicast route to the Local Domain. Those |
| 1419 |
border routers would be acting as de facto route controllers. This |
| 1420 |
approach would be, however, operationally cumbersome in a Local |
| 1421 |
Domain with numerous border routers having complex BGP policies.<a href="#section-1-2" class="pilcrow">¶</a></p> |
| 1422 |
<p id="section-1-3"> |
| 1423 |
<a href="#fig_1" class="xref">Figure 1</a> illustrates this principle. R1 (the upstream router) and |
| 1424 |
RR (a route reflector) need to validate the Flow Specification |
| 1425 |
whose embedded destination prefix has a best-match unicast route |
| 1426 |
(dest-route) originated by ASBR2. ASBR2 could originate the Flow |
| 1427 |
Specification, and it would be validated when received by RR and R1 |
| 1428 |
(from their point of view, the originator of both the Flow |
| 1429 |
Specification and the best-match unicast route will be ASBR1). |
| 1430 |
Sometimes the Flow Specification needs to be originated within AS1. |
| 1431 |
ASBR1 could originate it, and the Flow Specification would still be |
| 1432 |
validated. In both cases, the Flow Specification is originated by |
| 1433 |
a router in the same forwarding path as the dest-route. For the |
| 1434 |
case where AS1 has thousands of ASBRs, it becomes impractical to |
| 1435 |
originate different Flow Specification rules on each ASBR in AS1 |
| 1436 |
based on which ASBR each dest-route is learned from. To make the |
| 1437 |
situation more tenable, the objective is to advertise all the Flow |
| 1438 |
Specifications from the same route controller.<a href="#section-1-3" class="pilcrow">¶</a></p> |
| 1439 |
<div id="fig_1"> |
| 1440 |
<figure id="figure-1"> |
| 1441 |
<div class="artwork art-text alignLeft" id="section-1-4.1"> |
| 1442 |
<pre> |
| 1443 |
R1(AS1) --- RR(AS1) --- ASBR1(AS1) --- ASBR2(AS2) |
| 1444 |
| |
| 1445 |
route controller(AS1) |
| 1446 |
</pre> |
| 1447 |
</div> |
| 1448 |
<figcaption><a href="#figure-1" class="selfRef">Figure 1</a></figcaption></figure> |
| 1449 |
</div> |
| 1450 |
<p id="section-1-5"> This document describes a modification to the validation procedure described in <span>[<a href="#RFC8955" class="xref">RFC8955</a>]</span>, by allowing Flow Specification |
| 1451 |
NLRIs to be originated from a centralized BGP route controller located |
| 1452 |
within the Local Domain and not necessarily in the data-forwarding path. |
| 1453 |
While the proposed modification cannot be used for inter-domain |
| 1454 |
coordination of traffic filtering, it greatly simplifies distribution of |
| 1455 |
intra-domain traffic filtering policies within a Local Domain that has |
| 1456 |
numerous border routers having complex BGP policies. By relaxing the |
| 1457 |
validation procedure for iBGP, the proposed modification allows Flow |
| 1458 |
Specifications to be distributed in a standard and scalable manner |
| 1459 |
throughout the Local Domain.<a href="#section-1-5" class="pilcrow">¶</a></p> |
| 1460 |
<p id="section-1-6"> |
| 1461 |
|
| 1462 |
Throughout this document, some references are made to |
| 1463 |
AS_CONFED_SEQUENCE segments; see Sections <a href="#REV_ROUTE" class="xref">4.1</a> and <a href="#topology" class="xref">5</a>. If AS_CONFED_SET |
| 1464 |
segments are also present in the AS_PATH, the same |
| 1465 |
considerations apply to them. Note, however, that the use of |
| 1466 |
AS_CONFED_SET segments is not recommended <span>[<a href="#RFC6472" class="xref">RFC6472</a>]</span>. Refer to <span>[<a href="#I-D.ietf-idr-deprecate-as-set-confed-set" class="xref">CONFED-SET</a>]</span> as well.<a href="#section-1-6" class="pilcrow">¶</a></p> |
| 1467 |
</section> |
| 1468 |
<section id="section-2"> |
| 1469 |
<h2 id="name-definitions-of-terms-used-i"> |
| 1470 |
<a href="#section-2" class="section-number selfRef">2. </a><a href="#name-definitions-of-terms-used-i" class="section-name selfRef">Definitions of Terms Used in This Memo</a> |
| 1471 |
</h2> |
| 1472 |
<span class="break"></span><dl class="dlParallel" id="section-2-1"> |
| 1473 |
<dt id="section-2-1.1">Local Domain: |
| 1474 |
</dt> |
| 1475 |
<dd style="margin-left: 1.5em" id="section-2-1.2">the local AS or the local confederation of ASes <span>[<a href="#RFC5065" class="xref">RFC5065</a>]</span>.<a href="#section-2-1.2" class="pilcrow">¶</a> |
| 1476 |
</dd> |
| 1477 |
<dd class="break"></dd> |
| 1478 |
<dt id="section-2-1.3">eBGP: |
| 1479 |
</dt> |
| 1480 |
<dd style="margin-left: 1.5em" id="section-2-1.4">BGP peering to a router not within the Local Domain.<a href="#section-2-1.4" class="pilcrow">¶</a> |
| 1481 |
</dd> |
| 1482 |
<dd class="break"></dd> |
| 1483 |
<dt id="section-2-1.5">iBGP: |
| 1484 |
</dt> |
| 1485 |
<dd style="margin-left: 1.5em" id="section-2-1.6">Both classic iBGP and any form of eBGP peering with a router within the |
| 1486 |
same confederation (i.e., iBGP peering is a peering that is not eBGP as |
| 1487 |
defined above).<a href="#section-2-1.6" class="pilcrow">¶</a> |
| 1488 |
</dd> |
| 1489 |
<dd class="break"></dd> |
| 1490 |
</dl> |
| 1491 |
<p id="section-2-2"> |
| 1492 |
The key words "<span class="bcp14">MUST</span>", "<span class="bcp14">MUST NOT</span>", "<span class="bcp14">REQUIRED</span>", "<span class="bcp14">SHALL</span>", "<span class="bcp14">SHALL NOT</span>", "<span class="bcp14">SHOULD</span>", "<span class="bcp14">SHOULD NOT</span>", "<span class="bcp14">RECOMMENDED</span>", "<span class="bcp14">NOT RECOMMENDED</span>", |
| 1493 |
"<span class="bcp14">MAY</span>", and "<span class="bcp14">OPTIONAL</span>" in this document are to be interpreted as |
| 1494 |
described in BCP 14 <span>[<a href="#RFC2119" class="xref">RFC2119</a>]</span> <span>[<a href="#RFC8174" class="xref">RFC8174</a>]</span> |
| 1495 |
when, and only when, they appear in all capitals, as shown here.<a href="#section-2-2" class="pilcrow">¶</a></p> |
| 1496 |
</section> |
| 1497 |
<section id="section-3"> |
| 1498 |
<h2 id="name-motivation"> |
| 1499 |
<a href="#section-3" class="section-number selfRef">3. </a><a href="#name-motivation" class="section-name selfRef">Motivation</a> |
| 1500 |
</h2> |
| 1501 |
<p id="section-3-1">Step (b) of the validation procedure in <span><a href="https://www.rfc-editor.org/rfc/rfc8955#section-6" class="relref">Section 6</a> of [<a href="#RFC8955" class="xref">RFC8955</a>]</span> is defined with the |
| 1502 |
underlying assumption that the Flow Specification NLRI traverses the |
| 1503 |
same path, in the inter-domain and intra-domain route distribution |
| 1504 |
graph, as that of the longest-match unicast route for the destination |
| 1505 |
prefix embedded in the Flow Specification.<a href="#section-3-1" class="pilcrow">¶</a></p> |
| 1506 |
<p id="section-3-2"> |
| 1507 |
In the case of inter-domain traffic filtering, the Flow Specification |
| 1508 |
originator at the egress border routers of an AS (e.g., RTR-D and RTR-E |
| 1509 |
of AS1 in <a href="#fig_2" class="xref">Figure 2</a>) matches the eBGP neighbor that |
| 1510 |
advertised the longest match destination prefix (see RTR-F and RTR-G, |
| 1511 |
respectively, in <a href="#fig_2" class="xref">Figure 2</a>).<a href="#section-3-2" class="pilcrow">¶</a></p> |
| 1512 |
<p id="section-3-3"> |
| 1513 |
Similarly, at the upstream routers of an AS (see RTR-A and RTR-B of |
| 1514 |
AS1 in <a href="#fig_2" class="xref">Figure 2</a>), the Flow Specification originator |
| 1515 |
matches the egress iBGP border routers that had advertised the |
| 1516 |
unicast route for the best-match destination prefix (see RTR-D and |
| 1517 |
RTR-E, respectively, in <a href="#fig_2" class="xref">Figure 2</a>). This is true even |
| 1518 |
when upstream routers select paths from different egress border |
| 1519 |
routers as the best route based upon IGP distance. For example, in <a href="#fig_2" class="xref">Figure 2</a>:<a href="#section-3-3" class="pilcrow">¶</a></p> |
| 1520 |
<ul class="normal ulEmpty"> |
| 1521 |
<li class="normal ulEmpty" id="section-3-4.1">RTR-A chooses RTR-D as the best route<a href="#section-3-4.1" class="pilcrow">¶</a> |
| 1522 |
</li> |
| 1523 |
<li class="normal ulEmpty" id="section-3-4.2">RTR-B chooses RTR-E as the best route<a href="#section-3-4.2" class="pilcrow">¶</a> |
| 1524 |
</li> |
| 1525 |
</ul> |
| 1526 |
<div id="fig_2"> |
| 1527 |
<figure id="figure-2"> |
| 1528 |
<div class="artwork art-text alignLeft" id="section-3-5.1"> |
| 1529 |
<pre> |
| 1530 |
/ - - - - - - - - - - - - - - |
| 1531 |
| AS1 | |
| 1532 |
+-------+ +-------+ |
| 1533 |
| | | | | | |
| 1534 |
| RTR-A | | RTR-B | |
| 1535 |
| | | | | | |
| 1536 |
+-------+ +-------+ |
| 1537 |
| \ / | |
| 1538 |
iBGP \ / iBGP |
| 1539 |
| \ / | |
| 1540 |
+-------+ |
| 1541 |
| | | | |
| 1542 |
| RTR-C | |
| 1543 |
| | RC | | |
| 1544 |
+-------+ |
| 1545 |
| / \ | |
| 1546 |
/ \ |
| 1547 |
| iBGP / \ iBGP | |
| 1548 |
+-------+ +-------+ |
| 1549 |
| | RTR-D | | RTR-E | | |
| 1550 |
| | | | |
| 1551 |
| | | | | | |
| 1552 |
+-------+ +-------+ |
| 1553 |
| | | | |
| 1554 |
- - -|- - - - - - - - -|- - -/ |
| 1555 |
| eBGP eBGP | |
| 1556 |
- - -|- - - - - - - - -|- - -/ |
| 1557 |
| | | | |
| 1558 |
+-------+ +-------+ |
| 1559 |
| | | | | | |
| 1560 |
| RTR-F | | RTR-G | |
| 1561 |
| | | | | | |
| 1562 |
+-------+ +-------+ |
| 1563 |
| AS2 | |
| 1564 |
/ - - - - - - - - - - - - - - |
| 1565 |
</pre> |
| 1566 |
</div> |
| 1567 |
<figcaption><a href="#figure-2" class="selfRef">Figure 2</a></figcaption></figure> |
| 1568 |
</div> |
| 1569 |
<p id="section-3-6">It is highly desirable that mechanisms exist to protect each AS independently |
| 1570 |
from network security attacks using the BGP Flow Specification NLRI for |
| 1571 |
intra-AS purposes only. Network operators often deploy a dedicated |
| 1572 |
Security Operations Center (SOC) within their AS to monitor and detect such security attacks. |
| 1573 |
To mitigate attacks within an AS, operators require |
| 1574 |
the ability to originate intra-AS Flow Specification NLRIs from a |
| 1575 |
central BGP route controller that is not within the data forwarding plane. |
| 1576 |
In this way, operators can direct border routers within their AS with |
| 1577 |
specific attack-mitigation actions (drop the traffic, forward to a pipe-cleaning location, etc.).<a href="#section-3-6" class="pilcrow">¶</a></p> |
| 1578 |
<p id="section-3-7"> |
| 1579 |
In addition, an operator may extend the requirements above for a group of |
| 1580 |
ASes via policy. This is described in <a href="#REV_ROUTE" class="xref">Section 4.1</a> (<a href="#b.2.3" class="xref">b.2.3</a>) of the validation procedure.<a href="#section-3-7" class="pilcrow">¶</a></p> |
| 1581 |
<p id="section-3-8"> |
| 1582 |
A central BGP route controller that originates Flow Specification |
| 1583 |
NLRI should be able to avoid the complexity of having to determine |
| 1584 |
the egress border router whose path was chosen as the best for each |
| 1585 |
of its neighbors. |
| 1586 |
|
| 1587 |
When a central BGP route controller originates Flow Specification NLRI, the rest of the speakers |
| 1588 |
within the AS will see the BGP route controller as the originator of the Flow Specification in terms |
| 1589 |
of the validation procedure rules. Thus, it is necessary to modify step (b) of the validation procedure described in <span>[<a href="#RFC8955" class="xref">RFC8955</a>]</span> |
| 1590 |
such that an iBGP peer that is not within the data forwarding plane |
| 1591 |
may originate Flow Specification NLRIs.<a href="#section-3-8" class="pilcrow">¶</a></p> |
| 1592 |
</section> |
| 1593 |
<section id="section-4"> |
| 1594 |
<h2 id="name-revised-validation-procedur"> |
| 1595 |
<a href="#section-4" class="section-number selfRef">4. </a><a href="#name-revised-validation-procedur" class="section-name selfRef">Revised Validation Procedure</a> |
| 1596 |
</h2> |
| 1597 |
<div id="REV_ROUTE"> |
| 1598 |
<section id="section-4.1"> |
| 1599 |
<h3 id="name-revision-of-route-feasibili"> |
| 1600 |
<a href="#section-4.1" class="section-number selfRef">4.1. </a><a href="#name-revision-of-route-feasibili" class="section-name selfRef">Revision of Route Feasibility</a> |
| 1601 |
</h3> |
| 1602 |
<p id="section-4.1-1">Step (b) of the validation procedure specified in <span><a href="https://www.rfc-editor.org/rfc/rfc8955#section-6" class="relref">Section 6</a> of [<a href="#RFC8955" class="xref">RFC8955</a>]</span> is |
| 1603 |
redefined as follows:<a href="#section-4.1-1" class="pilcrow">¶</a></p> |
| 1604 |
<blockquote id="section-4.1-2"> |
| 1605 |
<span class="break"></span><dl class="dlParallel" id="section-4.1-2.1"> |
| 1606 |
<dt id="section-4.1-2.1.1">b)</dt> |
| 1607 |
<dd style="margin-left: 1.5em" id="section-4.1-2.1.2"> |
| 1608 |
<p id="section-4.1-2.1.2.1">One of the following conditions <span class="bcp14">MUST</span> hold true:<a href="#section-4.1-2.1.2.1" class="pilcrow">¶</a></p> |
| 1609 |
<div id="step_b"> |
| 1610 |
<ol start="1" type="1" class="normal type-1" id="section-4.1-2.1.2.2"> |
| 1611 |
<li id="section-4.1-2.1.2.2.1"> |
| 1612 |
<div id="b.1">The originator of the Flow Specification matches the |
| 1613 |
originator of the best-match unicast route for the destination |
| 1614 |
prefix embedded in the Flow Specification (this is the unicast |
| 1615 |
route with the longest possible prefix length covering the |
| 1616 |
destination prefix embedded in the Flow Specification).<a href="#b.1" class="pilcrow">¶</a> |
| 1617 |
</div> |
| 1618 |
</li> |
| 1619 |
<li id="section-4.1-2.1.2.2.2"> |
| 1620 |
<div id="b.2"> |
| 1621 |
<p id="section-4.1-2.1.2.2.2.1">The AS_PATH attribute of the Flow Specification is empty or |
| 1622 |
contains only an AS_CONFED_SEQUENCE segment <span>[<a href="#RFC5065" class="xref">RFC5065</a>]</span>.<a href="#section-4.1-2.1.2.2.2.1" class="pilcrow">¶</a></p> |
| 1623 |
<ol start="1" type="1" class="normal type-1" id="section-4.1-2.1.2.2.2.2"> |
| 1624 |
<li id="section-4.1-2.1.2.2.2.2.1"> |
| 1625 |
<div id="b.1.1">This condition <span class="bcp14">SHOULD</span> be |
| 1626 |
enabled by default.<a href="#b.1.1" class="pilcrow">¶</a> |
| 1627 |
</div> |
| 1628 |
</li> |
| 1629 |
<li id="section-4.1-2.1.2.2.2.2.2"> |
| 1630 |
<div id="b.2.2">This condition <span class="bcp14">MAY</span> be disabled by |
| 1631 |
explicit configuration on a BGP speaker.<a href="#b.2.2" class="pilcrow">¶</a> |
| 1632 |
</div> |
| 1633 |
</li> |
| 1634 |
<li id="section-4.1-2.1.2.2.2.2.3"> |
| 1635 |
<div id="b.2.3">As an extension to this rule, a given non-empty AS_PATH |
| 1636 |
(besides AS_CONFED_SEQUENCE segments) <span class="bcp14">MAY</span> be |
| 1637 |
permitted by policy.<a href="#b.2.3" class="pilcrow">¶</a> |
| 1638 |
</div> |
| 1639 |
</li> |
| 1640 |
</ol> |
| 1641 |
</div> |
| 1642 |
</li> |
| 1643 |
</ol> |
| 1644 |
</div> |
| 1645 |
</dd> |
| 1646 |
<dd class="break"></dd> |
| 1647 |
</dl> |
| 1648 |
</blockquote> |
| 1649 |
<p id="section-4.1-3">Explanation:<a href="#section-4.1-3" class="pilcrow">¶</a></p> |
| 1650 |
<ul class="normal ulEmpty"> |
| 1651 |
<li class="normal ulEmpty" id="section-4.1-4.1"> |
| 1652 |
Receiving either an empty AS_PATH or one |
| 1653 |
with only an AS_CONFED_SEQUENCE segment indicates that the Flow Specification was |
| 1654 |
originated inside the Local Domain.<a href="#section-4.1-4.1" class="pilcrow">¶</a> |
| 1655 |
</li> |
| 1656 |
<li class="normal ulEmpty" id="section-4.1-4.2"> |
| 1657 |
With the above modification to the <span>[<a href="#RFC8955" class="xref">RFC8955</a>]</span> validation procedure, a BGP peer within the Local Domain |
| 1658 |
that is not within the data-forwarding path can originate a Flow Specification.<a href="#section-4.1-4.2" class="pilcrow">¶</a> |
| 1659 |
</li> |
| 1660 |
<li class="normal ulEmpty" id="section-4.1-4.3"> |
| 1661 |
Disabling the new condition above (see <a href="#b.2.2" class="xref">step |
| 1662 |
b.2.2</a> in <a href="#REV_ROUTE" class="xref">Section 4.1</a>) could be a good practice if the |
| 1663 |
operator knew with certainty that a Flow Specification would not be originated |
| 1664 |
inside the Local Domain. An additional case would be if it was known for a |
| 1665 |
fact that only the right egress border routers (i.e., those that were also |
| 1666 |
egress border routers for the best routes) were originating Flow Specification |
| 1667 |
NLRI.<a href="#section-4.1-4.3" class="pilcrow">¶</a> |
| 1668 |
</li> |
| 1669 |
<li class="normal ulEmpty" id="section-4.1-4.4"> |
| 1670 |
Also, policy may be useful to permit a specific set of non-empty AS_PATHs (see |
| 1671 |
<a href="#b.2.3" class="xref">step b.2.3</a> in <a href="#REV_ROUTE" class="xref">Section 4.1</a>). For example, it could validate a Flow Specification |
| 1672 |
whose AS_PATH contained only an AS_SEQUENCE segment with ASes that were all |
| 1673 |
known to belong to the same administrative domain.<a href="#section-4.1-4.4" class="pilcrow">¶</a> |
| 1674 |
</li> |
| 1675 |
</ul> |
| 1676 |
</section> |
| 1677 |
</div> |
| 1678 |
<div id="AS_PATH"> |
| 1679 |
<section id="section-4.2"> |
| 1680 |
<h3 id="name-revision-of-as_path-validat"> |
| 1681 |
<a href="#section-4.2" class="section-number selfRef">4.2. </a><a href="#name-revision-of-as_path-validat" class="section-name selfRef">Revision of AS_PATH Validation</a> |
| 1682 |
</h3> |
| 1683 |
<p id="section-4.2-1"> |
| 1684 |
<span><a href="https://www.rfc-editor.org/rfc/rfc8955#section-6" class="relref">Section 6</a> of [<a href="#RFC8955" class="xref">RFC8955</a>]</span> |
| 1685 |
states:<a href="#section-4.2-1" class="pilcrow">¶</a></p> |
| 1686 |
<ul class="normal ulEmpty"> |
| 1687 |
<li class="normal ulEmpty" id="section-4.2-2.1"> |
| 1688 |
<blockquote id="section-4.2-2.1.1"> |
| 1689 |
BGP implementations <span class="bcp14">MUST</span> also enforce that the |
| 1690 |
AS_PATH attribute of a route received via the External Border Gateway Protocol (eBGP) |
| 1691 |
contains the neighboring AS in the left-most position of the AS_PATH attribute. While this rule is optional in the BGP specification, it |
| 1692 |
becomes necessary to enforce it here for security reasons.<a href="#section-4.2-2.1.1" class="pilcrow">¶</a> |
| 1693 |
</blockquote> |
| 1694 |
</li> |
| 1695 |
</ul> |
| 1696 |
<p id="section-4.2-3"> |
| 1697 |
This rule prevents the exchange of BGP Flow Specification NLRIs at Internet |
| 1698 |
exchanges with BGP route servers, which by design don't insert their own AS |
| 1699 |
number into the AS_PATH (<span><a href="https://www.rfc-editor.org/rfc/rfc7947#section-2.2.2.1" class="relref">Section 2.2.2.1</a> of [<a href="#RFC7947" class="xref">RFC7947</a>]</span>). Therefore, this document also |
| 1700 |
redefines the <span>[<a href="#RFC8955" class="xref">RFC8955</a>]</span> AS_PATH validation |
| 1701 |
procedure referenced above as follows:<a href="#section-4.2-3" class="pilcrow">¶</a></p> |
| 1702 |
<ul class="normal ulEmpty"> |
| 1703 |
<li class="normal ulEmpty" id="section-4.2-4.1"> |
| 1704 |
<blockquote id="section-4.2-4.1.1"> |
| 1705 |
BGP Flow Specification implementations <span class="bcp14">MUST</span> enforce that the AS in the left-most position of the AS_PATH attribute of a Flow Specification route |
| 1706 |
received via the External Border Gateway Protocol (eBGP) matches the AS in the left-most position of the AS_PATH attribute of the best-match unicast route for the destination prefix |
| 1707 |
embedded in the Flow Specification NLRI.<a href="#section-4.2-4.1.1" class="pilcrow">¶</a> |
| 1708 |
</blockquote> |
| 1709 |
</li> |
| 1710 |
</ul> |
| 1711 |
<p id="section-4.2-5"> |
| 1712 |
Explanation:<a href="#section-4.2-5" class="pilcrow">¶</a></p> |
| 1713 |
<ul class="normal ulEmpty"> |
| 1714 |
<li class="normal ulEmpty" id="section-4.2-6.1">For clarity, the AS in the left-most position of the AS_PATH means the AS that was last added to an AS_SEQUENCE.<a href="#section-4.2-6.1" class="pilcrow">¶</a> |
| 1715 |
</li> |
| 1716 |
<li class="normal ulEmpty" id="section-4.2-6.2">This proposed modification enables the exchange of |
| 1717 |
BGP Flow Specification NLRIs at Internet exchanges with |
| 1718 |
BGP route servers while at the same time, for security reasons, |
| 1719 |
prevents an eBGP peer from advertising an inter-domain |
| 1720 |
Flow Specification for a destination prefix that it does |
| 1721 |
not provide reachability information for.<a href="#section-4.2-6.2" class="pilcrow">¶</a> |
| 1722 |
</li> |
| 1723 |
<li class="normal ulEmpty" id="section-4.2-6.3"> |
| 1724 |
Comparing only the left-most AS in the AS-PATH for eBGP-learned Flow Specification NLRIs is |
| 1725 |
roughly equivalent to checking the neighboring AS. |
| 1726 |
If the peer is a route server, security is necessarily weakened for the Flow Specification NLRI, as it is for any unicast route advertised from a route server. An example is discussed in the <a href="#Security" class="xref">Security Considerations</a> section.<a href="#section-4.2-6.3" class="pilcrow">¶</a> |
| 1727 |
</li> |
| 1728 |
<li class="normal ulEmpty" id="section-4.2-6.4"> |
| 1729 |
Redefinition of this AS_PATH validation rule for a Flow Specification does not |
| 1730 |
mean that the original rule in <span>[<a href="#RFC8955" class="xref">RFC8955</a>]</span> |
| 1731 |
cannot be enforced as well. Its enforcement remains optional per <span><a href="https://www.rfc-editor.org/rfc/rfc4271#section-6.3" class="relref">Section 6.3</a> of [<a href="#RFC4271" class="xref">RFC4271</a>]</span>. That |
| 1732 |
is, a BGP speaker can enforce the first AS in the AS_PATH to be the same as |
| 1733 |
the neighbor AS for a route belonging to any Address Family (including Flow |
| 1734 |
Specification Address Family). If the BGP speaker peer is not a route server, |
| 1735 |
when enforcing this optional rule, the security characteristics are exactly |
| 1736 |
equivalent to those specified in <span>[<a href="#RFC8955" class="xref">RFC8955</a>]</span>.<a href="#section-4.2-6.4" class="pilcrow">¶</a> |
| 1737 |
</li> |
| 1738 |
<li class="normal ulEmpty" id="section-4.2-6.5"> |
| 1739 |
Alternatively, enforcing this optional rule for unicast routes (even if not enforced on Flow Specification NLRIs) achieves exactly the same security characteristics. |
| 1740 |
The reason is that, after all validations, the neighboring AS will be the same as the left-most AS in the AS-PATH for the unicast route, and the left-most AS in the AS_PATH for the unicast route |
| 1741 |
will be the same as the left-most AS in the AS_PATH for the Flow Specification NLRI. Therefore, the neighboring AS will be the same as the left-most AS in the AS_PATH for the Flow Specification NLRI (as the original |
| 1742 |
AS_PATH validation rule in <span>[<a href="#RFC8955" class="xref">RFC8955</a>]</span> states).<a href="#section-4.2-6.5" class="pilcrow">¶</a> |
| 1743 |
</li> |
| 1744 |
<li class="normal ulEmpty" id="section-4.2-6.6"> |
| 1745 |
Note, however, that not checking the full AS_PATH allows any rogue or |
| 1746 |
misconfigured AS the ability to originate undesired Flow Specifications. This |
| 1747 |
is a BGP security threat, already present in <span>[<a href="#RFC8955" class="xref">RFC8955</a>]</span>, but out of the scope of this document.<a href="#section-4.2-6.6" class="pilcrow">¶</a> |
| 1748 |
</li> |
| 1749 |
<li class="normal ulEmpty" id="section-4.2-6.7"> |
| 1750 |
Using the new rule to validate a Flow Specification route received from a peer belonging to the same Local Domain |
| 1751 |
is out of the scope of this document. Note that although it's possible, its utility is dubious. |
| 1752 |
Although it is conceivable that a router in the same Local Domain could send a rogue update, only eBGP risk is considered within this document |
| 1753 |
(in the same spirit as the aforementioned AS_PATH validation in <span>[<a href="#RFC4271" class="xref">RFC4271</a>]</span>).<a href="#section-4.2-6.7" class="pilcrow">¶</a> |
| 1754 |
</li> |
| 1755 |
</ul> |
| 1756 |
</section> |
| 1757 |
</div> |
| 1758 |
</section> |
| 1759 |
<div id="topology"> |
| 1760 |
<section id="section-5"> |
| 1761 |
<h2 id="name-topology-considerations"> |
| 1762 |
<a href="#section-5" class="section-number selfRef">5. </a><a href="#name-topology-considerations" class="section-name selfRef">Topology Considerations</a> |
| 1763 |
</h2> |
| 1764 |
<p id="section-5-1"> |
| 1765 |
<span>[<a href="#RFC8955" class="xref">RFC8955</a>]</span> indicates that the originator may |
| 1766 |
refer to the originator path attribute (ORIGINATOR_ID) or (if the attribute is |
| 1767 |
not present) the transport address of the peer from which the BGP speaker |
| 1768 |
received the update. If the latter applies, a network should be designed so |
| 1769 |
it has a congruent topology amongst unicast routes and Flow Specification |
| 1770 |
routes. By congruent topology, it is understood that the two routes (i.e., |
| 1771 |
the Flow Specification route and its best-match unicast route) are learned |
| 1772 |
from the same peer across the AS. That would likely not be true, for |
| 1773 |
instance, if some peers only negotiated one Address Family or if each Address |
| 1774 |
Family peering had a different set of policies. Failing to have a congruent |
| 1775 |
topology would result in step (<a href="#b.1" class="xref">b.1</a>) of the |
| 1776 |
validation procedure to fail.<a href="#section-5-1" class="pilcrow">¶</a></p> |
| 1777 |
<p id="section-5-2"> |
| 1778 |
With the additional second condition (<a href="#b.2" class="xref">b.2</a>) in the validation procedure, non-congruent topologies are supported within the Local Domain if the Flow Specification |
| 1779 |
is originated within the Local Domain.<a href="#section-5-2" class="pilcrow">¶</a></p> |
| 1780 |
<p id="section-5-3"> |
| 1781 |
Explanation:<a href="#section-5-3" class="pilcrow">¶</a></p> |
| 1782 |
<ul class="normal ulEmpty"> |
| 1783 |
<li class="normal ulEmpty" id="section-5-4.1"> |
| 1784 |
<p id="section-5-4.1.1">Consider the following scenarios of a non-congruent topology without the second condition (<a href="#b.2" class="xref">b.2</a>) being added to the validation procedure:<a href="#section-5-4.1.1" class="pilcrow">¶</a></p> |
| 1785 |
<ol start="1" type="1" class="normal type-1" id="section-5-4.1.2"> |
| 1786 |
<li id="section-5-4.1.2.1">Consider a topology with two BGP |
| 1787 |
speakers with two iBGP peering sessions between them, one for |
| 1788 |
unicast and one for Flow Specification. This is a non-congruent |
| 1789 |
topology. Let's assume that the ORIGINATOR_ID attribute was not |
| 1790 |
received (e.g., a route reflector receiving routes from its |
| 1791 |
clients). In this case, the Flow Specification validation procedure |
| 1792 |
will fail because of the first condition (<a href="#b.1" class="xref">b.1</a>).<a href="#section-5-4.1.2.1" class="pilcrow">¶</a> |
| 1793 |
</li> |
| 1794 |
<li id="section-5-4.1.2.2">Consider a confederation of ASes with local AS X and local AS Y (both belonging to the same Local Domain), and a given BGP speaker X1 inside local AS X. |
| 1795 |
The ORIGINATOR_ID attribute is not advertised when propagating routes across local ASes. |
| 1796 |
Let's assume the Flow Specification route is received from peer Y1 and the best-match unicast route |
| 1797 |
is received from peer Y2. Both peers belong to local AS Y. |
| 1798 |
The Flow Specification validation procedure will also fail because of the first condition (<a href="#b.1" class="xref">b.1</a>).<a href="#section-5-4.1.2.2" class="pilcrow">¶</a> |
| 1799 |
</li> |
| 1800 |
</ol> |
| 1801 |
</li> |
| 1802 |
<li class="normal ulEmpty" id="section-5-4.2"> |
| 1803 |
Consider now that the second condition (<a href="#b.2" class="xref">b.2</a>) is |
| 1804 |
added to the validation procedure. In the scenarios above, if Flow |
| 1805 |
Specifications are originated in the same Local Domain, the AS_PATH will be |
| 1806 |
empty or contain only an AS_CONFED_SEQUENCE segment. Condition (<a href="#b.2" class="xref">b.2</a>) will evaluate to true. Therefore, using the second |
| 1807 |
condition (<a href="#b.2" class="xref">b.2</a>), as defined by this document, |
| 1808 |
guarantees that the overall validation procedure will pass. Thus, |
| 1809 |
non-congruent topologies are supported if the Flow Specification is originated |
| 1810 |
in the same Local Domain.<a href="#section-5-4.2" class="pilcrow">¶</a> |
| 1811 |
</li> |
| 1812 |
<li class="normal ulEmpty" id="section-5-4.3"> |
| 1813 |
Flow Specifications originated in a different Local Domain sill need a |
| 1814 |
congruent topology. The reason is that in a non-congruent topology, the second |
| 1815 |
condition (<a href="#b.2" class="xref">b.2</a>) evaluates to false and |
| 1816 |
only the first condition (<a href="#b.1" class="xref">b.1</a>) is |
| 1817 |
evaluated.<a href="#section-5-4.3" class="pilcrow">¶</a> |
| 1818 |
</li> |
| 1819 |
</ul> |
| 1820 |
</section> |
| 1821 |
</div> |
| 1822 |
<div id="IANA"> |
| 1823 |
<section id="section-6"> |
| 1824 |
<h2 id="name-iana-considerations"> |
| 1825 |
<a href="#section-6" class="section-number selfRef">6. </a><a href="#name-iana-considerations" class="section-name selfRef">IANA Considerations</a> |
| 1826 |
</h2> |
| 1827 |
<p id="section-6-1">This document has no IANA actions.<a href="#section-6-1" class="pilcrow">¶</a></p> |
| 1828 |
</section> |
| 1829 |
</div> |
| 1830 |
<div id="Security"> |
| 1831 |
<section id="section-7"> |
| 1832 |
<h2 id="name-security-considerations"> |
| 1833 |
<a href="#section-7" class="section-number selfRef">7. </a><a href="#name-security-considerations" class="section-name selfRef">Security Considerations</a> |
| 1834 |
</h2> |
| 1835 |
<p id="section-7-1"> |
| 1836 |
This document updates the route feasibility validation procedures for Flow |
| 1837 |
Specifications learned from iBGP peers and through route servers. This |
| 1838 |
change is in line with the procedures described in <span>[<a href="#RFC8955" class="xref">RFC8955</a>]</span> and, thus, security characteristics remain essentially |
| 1839 |
equivalent to the existing security properties of BGP unicast routing, |
| 1840 |
except as detailed below.<a href="#section-7-1" class="pilcrow">¶</a></p> |
| 1841 |
<p id="section-7-2"> |
| 1842 |
The security considerations discussed in <span>[<a href="#RFC8955" class="xref">RFC8955</a>]</span> apply to this |
| 1843 |
specification as well.<a href="#section-7-2" class="pilcrow">¶</a></p> |
| 1844 |
<p id="section-7-3"> |
| 1845 |
This document makes the original AS_PATH validation rule (<span><a href="https://www.rfc-editor.org/rfc/rfc4271#section-6.3" class="relref">Section 6.3</a> of [<a href="#RFC4271" class="xref">RFC4271</a>]</span>) again |
| 1846 |
<span class="bcp14">OPTIONAL</span> (<a href="#AS_PATH" class="xref">Section 4.2</a>) for Flow Specification Address Family (the rule is no longer |
| 1847 |
mandatory as had been specified by <span>[<a href="#RFC8955" class="xref">RFC8955</a>]</span>). If that original rule is |
| 1848 |
not enforced for Flow Specification, it may introduce some new security |
| 1849 |
risks. A speaker in AS X peering with a route server could advertise a |
| 1850 |
rogue Flow Specification route whose first AS in AS_PATH was Y. Assume Y is |
| 1851 |
the first AS in the AS_PATH of the best-match unicast route. When the |
| 1852 |
route server advertises the Flow Specification to a speaker in AS Z, it |
| 1853 |
will be validated by that speaker. This risk is impossible to prevent if |
| 1854 |
the Flow Specification route is received from a route server peer. If |
| 1855 |
configuration (or other means beyond the scope of this document) indicates |
| 1856 |
that the peer is not a route server, that optional rule |
| 1857 |
<span class="bcp14">SHOULD</span> be enforced for unicast and/or for Flow |
| 1858 |
Specification routes (as discussed in the <a href="#AS_PATH" class="xref">Revision of AS_PATH Validation</a> section, just |
| 1859 |
enforcing it in one of those Address Families is enough). If the indication |
| 1860 |
is that the peer is not a route server or there is no conclusive |
| 1861 |
indication, that optional rule <span class="bcp14">SHOULD NOT</span> be enforced.<a href="#section-7-3" class="pilcrow">¶</a></p> |
| 1862 |
<p id="section-7-4"> |
| 1863 |
A route server itself may be in a good position to enforce the AS_PATH validation rule described |
| 1864 |
in the previous paragraph. If it is known that a route server is not peering with any other route server, |
| 1865 |
it can enforce the AS_PATH validation rule across all its peers.<a href="#section-7-4" class="pilcrow">¶</a></p> |
| 1866 |
<p id="section-7-5"> |
| 1867 |
BGP updates learned from iBGP peers are considered |
| 1868 |
trusted, so the Traffic Flow Specifications contained in BGP updates |
| 1869 |
are also considered trusted. Therefore, it is not required to |
| 1870 |
validate that the originator of an intra-domain Traffic Flow |
| 1871 |
Specification matches the originator of the best-match unicast route |
| 1872 |
for the destination prefix embedded in that Flow Specification. Note that this trustworthiness consideration is not |
| 1873 |
absolute and the new possibility that an iBGP speaker could send a rogue Flow Specification is introduced.<a href="#section-7-5" class="pilcrow">¶</a></p> |
| 1874 |
<p id="section-7-6"> |
| 1875 |
The changes in <a href="#REV_ROUTE" class="xref">Section 4.1</a> don't affect the validation |
| 1876 |
procedures for eBGP-learned routes.<a href="#section-7-6" class="pilcrow">¶</a></p> |
| 1877 |
<p id="section-7-7"> |
| 1878 |
It's worth mentioning that allowing (or making operationally feasible) |
| 1879 |
Flow Specifications to originate within the Local Domain makes |
| 1880 |
the network overall more secure. Flow Specifications can be originated |
| 1881 |
more readily during attacks and improve the stability and security of |
| 1882 |
the network.<a href="#section-7-7" class="pilcrow">¶</a></p> |
| 1883 |
</section> |
| 1884 |
</div> |
| 1885 |
<section id="section-8"> |
| 1886 |
<h2 id="name-references"> |
| 1887 |
<a href="#section-8" class="section-number selfRef">8. </a><a href="#name-references" class="section-name selfRef">References</a> |
| 1888 |
</h2> |
| 1889 |
<section id="section-8.1"> |
| 1890 |
<h3 id="name-normative-references"> |
| 1891 |
<a href="#section-8.1" class="section-number selfRef">8.1. </a><a href="#name-normative-references" class="section-name selfRef">Normative References</a> |
| 1892 |
</h3> |
| 1893 |
<dl class="references"> |
| 1894 |
<dt id="RFC2119">[RFC2119]</dt> |
| 1895 |
<dd> |
| 1896 |
<span class="refAuthor">Bradner, S.</span>, <span class="refTitle">"Key words for use in RFCs to Indicate Requirement Levels"</span>, <span class="seriesInfo">BCP 14</span>, <span class="seriesInfo">RFC 2119</span>, <span class="seriesInfo">DOI 10.17487/RFC2119</span>, <time datetime="1997-03" class="refDate">March 1997</time>, <span><<a href="https://www.rfc-editor.org/info/rfc2119">https://www.rfc-editor.org/info/rfc2119</a>></span>. </dd> |
| 1897 |
<dd class="break"></dd> |
| 1898 |
<dt id="RFC4271">[RFC4271]</dt> |
| 1899 |
<dd> |
| 1900 |
<span class="refAuthor">Rekhter, Y., Ed.</span>, <span class="refAuthor">Li, T., Ed.</span>, and <span class="refAuthor">S. Hares, Ed.</span>, <span class="refTitle">"A Border Gateway Protocol 4 (BGP-4)"</span>, <span class="seriesInfo">RFC 4271</span>, <span class="seriesInfo">DOI 10.17487/RFC4271</span>, <time datetime="2006-01" class="refDate">January 2006</time>, <span><<a href="https://www.rfc-editor.org/info/rfc4271">https://www.rfc-editor.org/info/rfc4271</a>></span>. </dd> |
| 1901 |
<dd class="break"></dd> |
| 1902 |
<dt id="RFC4760">[RFC4760]</dt> |
| 1903 |
<dd> |
| 1904 |
<span class="refAuthor">Bates, T.</span>, <span class="refAuthor">Chandra, R.</span>, <span class="refAuthor">Katz, D.</span>, and <span class="refAuthor">Y. Rekhter</span>, <span class="refTitle">"Multiprotocol Extensions for BGP-4"</span>, <span class="seriesInfo">RFC 4760</span>, <span class="seriesInfo">DOI 10.17487/RFC4760</span>, <time datetime="2007-01" class="refDate">January 2007</time>, <span><<a href="https://www.rfc-editor.org/info/rfc4760">https://www.rfc-editor.org/info/rfc4760</a>></span>. </dd> |
| 1905 |
<dd class="break"></dd> |
| 1906 |
<dt id="RFC5065">[RFC5065]</dt> |
| 1907 |
<dd> |
| 1908 |
<span class="refAuthor">Traina, P.</span>, <span class="refAuthor">McPherson, D.</span>, and <span class="refAuthor">J. Scudder</span>, <span class="refTitle">"Autonomous System Confederations for BGP"</span>, <span class="seriesInfo">RFC 5065</span>, <span class="seriesInfo">DOI 10.17487/RFC5065</span>, <time datetime="2007-08" class="refDate">August 2007</time>, <span><<a href="https://www.rfc-editor.org/info/rfc5065">https://www.rfc-editor.org/info/rfc5065</a>></span>. </dd> |
| 1909 |
<dd class="break"></dd> |
| 1910 |
<dt id="RFC7947">[RFC7947]</dt> |
| 1911 |
<dd> |
| 1912 |
<span class="refAuthor">Jasinska, E.</span>, <span class="refAuthor">Hilliard, N.</span>, <span class="refAuthor">Raszuk, R.</span>, and <span class="refAuthor">N. Bakker</span>, <span class="refTitle">"Internet Exchange BGP Route Server"</span>, <span class="seriesInfo">RFC 7947</span>, <span class="seriesInfo">DOI 10.17487/RFC7947</span>, <time datetime="2016-09" class="refDate">September 2016</time>, <span><<a href="https://www.rfc-editor.org/info/rfc7947">https://www.rfc-editor.org/info/rfc7947</a>></span>. </dd> |
| 1913 |
<dd class="break"></dd> |
| 1914 |
<dt id="RFC8174">[RFC8174]</dt> |
| 1915 |
<dd> |
| 1916 |
<span class="refAuthor">Leiba, B.</span>, <span class="refTitle">"Ambiguity of Uppercase vs Lowercase in RFC 2119 Key Words"</span>, <span class="seriesInfo">BCP 14</span>, <span class="seriesInfo">RFC 8174</span>, <span class="seriesInfo">DOI 10.17487/RFC8174</span>, <time datetime="2017-05" class="refDate">May 2017</time>, <span><<a href="https://www.rfc-editor.org/info/rfc8174">https://www.rfc-editor.org/info/rfc8174</a>></span>. </dd> |
| 1917 |
<dd class="break"></dd> |
| 1918 |
<dt id="RFC8955">[RFC8955]</dt> |
| 1919 |
<dd> |
| 1920 |
<span class="refAuthor">Loibl, C.</span>, <span class="refAuthor">Hares, S.</span>, <span class="refAuthor">Raszuk, R.</span>, <span class="refAuthor">McPherson, D.</span>, and <span class="refAuthor">M. Bacher</span>, <span class="refTitle">"Dissemination of Flow Specification Rules"</span>, <span class="seriesInfo">RFC 8955</span>, <span class="seriesInfo">DOI 10.17487/RFC8955</span>, <time datetime="2020-12" class="refDate">December 2020</time>, <span><<a href="https://www.rfc-editor.org/info/rfc8955">https://www.rfc-editor.org/info/rfc8955</a>></span>. </dd> |
| 1921 |
<dd class="break"></dd> |
| 1922 |
</dl> |
| 1923 |
</section> |
| 1924 |
<section id="section-8.2"> |
| 1925 |
<h3 id="name-informative-references"> |
| 1926 |
<a href="#section-8.2" class="section-number selfRef">8.2. </a><a href="#name-informative-references" class="section-name selfRef">Informative References</a> |
| 1927 |
</h3> |
| 1928 |
<dl class="references"> |
| 1929 |
<dt id="I-D.ietf-idr-deprecate-as-set-confed-set">[CONFED-SET]</dt> |
| 1930 |
<dd> |
| 1931 |
<span class="refAuthor">Kumari, W.</span>, <span class="refAuthor">Sriram, K.</span>, <span class="refAuthor">Hannachi, L.</span>, and <span class="refAuthor">J. Haas</span>, <span class="refTitle">"Deprecation of AS_SET and AS_CONFED_SET in BGP"</span>, <span class="refContent">Work in Progress</span>, <span class="seriesInfo">Internet-Draft, draft-ietf-idr-deprecate-as-set-confed-set-05</span>, <time datetime="2021-03-12" class="refDate">12 March 2021</time>, <span><<a href="https://datatracker.ietf.org/doc/html/draft-ietf-idr-deprecate-as-set-confed-set-05">https://datatracker.ietf.org/doc/html/draft-ietf-idr-deprecate-as-set-confed-set-05</a>></span>. </dd> |
| 1932 |
<dd class="break"></dd> |
| 1933 |
<dt id="RFC6472">[RFC6472]</dt> |
| 1934 |
<dd> |
| 1935 |
<span class="refAuthor">Kumari, W.</span> and <span class="refAuthor">K. Sriram</span>, <span class="refTitle">"Recommendation for Not Using AS_SET and AS_CONFED_SET in BGP"</span>, <span class="seriesInfo">BCP 172</span>, <span class="seriesInfo">RFC 6472</span>, <span class="seriesInfo">DOI 10.17487/RFC6472</span>, <time datetime="2011-12" class="refDate">December 2011</time>, <span><<a href="https://www.rfc-editor.org/info/rfc6472">https://www.rfc-editor.org/info/rfc6472</a>></span>. </dd> |
| 1936 |
<dd class="break"></dd> |
| 1937 |
</dl> |
| 1938 |
</section> |
| 1939 |
</section> |
| 1940 |
<div id="Acknowledgements"> |
| 1941 |
<section id="appendix-A"> |
| 1942 |
<h2 id="name-acknowledgements"> |
| 1943 |
<a href="#name-acknowledgements" class="section-name selfRef">Acknowledgements</a> |
| 1944 |
</h2> |
| 1945 |
<p id="appendix-A-1">The authors would like to thank <span class="contact-name">Han Nguyen</span> for |
| 1946 |
his direction on this work as well as <span class="contact-name">Waqas Alam</span>, |
| 1947 |
<span class="contact-name">Keyur Patel</span>, <span class="contact-name">Robert Raszuk</span>, |
| 1948 |
<span class="contact-name">Eric Rosen</span>, <span class="contact-name">Shyam Sethuram</span>, |
| 1949 |
<span class="contact-name">Susan Hares</span>, <span class="contact-name">Alvaro Retana</span>, |
| 1950 |
and <span class="contact-name">John Scudder</span> for their review and comments.<a href="#appendix-A-1" class="pilcrow">¶</a></p> |
| 1951 |
</section> |
| 1952 |
</div> |
| 1953 |
<div id="authors-addresses"> |
| 1954 |
<section id="appendix-B"> |
| 1955 |
<h2 id="name-authors-addresses"> |
| 1956 |
<a href="#name-authors-addresses" class="section-name selfRef">Authors' Addresses</a> |
| 1957 |
</h2> |
| 1958 |
<address class="vcard"> |
| 1959 |
<div dir="auto" class="left"><span class="fn nameRole">James Uttaro</span></div> |
| 1960 |
<div dir="auto" class="left"><span class="org">AT&T</span></div> |
| 1961 |
<div dir="auto" class="left"><span class="street-address">200 S. Laurel Ave</span></div> |
| 1962 |
<div dir="auto" class="left"> |
| 1963 |
<span class="locality">Middletown</span>, <span class="region">NJ</span> <span class="postal-code">07748</span> |
| 1964 |
</div> |
| 1965 |
<div dir="auto" class="left"><span class="country-name">United States of America</span></div> |
| 1966 |
<div class="email"> |
| 1967 |
<span>Email:</span> |
| 1968 |
<a href="mailto:[email protected]" class="email">[email protected]</a> |
| 1969 |
</div> |
| 1970 |
</address> |
| 1971 |
<address class="vcard"> |
| 1972 |
<div dir="auto" class="left"><span class="fn nameRole">Juan Alcaide</span></div> |
| 1973 |
<div dir="auto" class="left"><span class="org">Cisco</span></div> |
| 1974 |
<div dir="auto" class="left"><span class="extended-address">Research Triangle Park</span></div> |
| 1975 |
<div dir="auto" class="left"><span class="street-address">7100 Kit Creek Road</span></div> |
| 1976 |
<div dir="auto" class="left"> |
| 1977 |
<span class="locality">Morrisville</span>, <span class="region">NC</span> <span class="postal-code">27709</span> |
| 1978 |
</div> |
| 1979 |
<div dir="auto" class="left"><span class="country-name">United States of America</span></div> |
| 1980 |
<div class="email"> |
| 1981 |
<span>Email:</span> |
| 1982 |
<a href="mailto:[email protected]" class="email">[email protected]</a> |
| 1983 |
</div> |
| 1984 |
</address> |
| 1985 |
<address class="vcard"> |
| 1986 |
<div dir="auto" class="left"><span class="fn nameRole">Clarence Filsfils</span></div> |
| 1987 |
<div dir="auto" class="left"><span class="org">Cisco</span></div> |
| 1988 |
<div class="email"> |
| 1989 |
<span>Email:</span> |
| 1990 |
<a href="mailto:[email protected]" class="email">[email protected]</a> |
| 1991 |
</div> |
| 1992 |
</address> |
| 1993 |
<address class="vcard"> |
| 1994 |
<div dir="auto" class="left"><span class="fn nameRole">David Smith</span></div> |
| 1995 |
<div dir="auto" class="left"><span class="org">Cisco</span></div> |
| 1996 |
<div dir="auto" class="left"><span class="street-address">111 Wood Ave South</span></div> |
| 1997 |
<div dir="auto" class="left"> |
| 1998 |
<span class="locality">Iselin</span>, <span class="region">NJ</span> <span class="postal-code">08830</span> |
| 1999 |
</div> |
| 2000 |
<div dir="auto" class="left"><span class="country-name">United States of America</span></div> |
| 2001 |
<div class="email"> |
| 2002 |
<span>Email:</span> |
| 2003 |
<a href="mailto:[email protected]" class="email">[email protected]</a> |
| 2004 |
</div> |
| 2005 |
</address> |
| 2006 |
<address class="vcard"> |
| 2007 |
<div dir="auto" class="left"><span class="fn nameRole">Pradosh Mohapatra</span></div> |
| 2008 |
<div dir="auto" class="left"><span class="org">Sproute Networks</span></div> |
| 2009 |
<div class="email"> |
| 2010 |
<span>Email:</span> |
| 2011 |
<a href="mailto:[email protected]" class="email">[email protected]</a> |
| 2012 |
</div> |
| 2013 |
</address> |
| 2014 |
</section> |
| 2015 |
</div> |
| 2016 |
<script>const toc = document.getElementById("toc"); |
| 2017 |
toc.querySelector("h2").addEventListener("click", e => { |
| 2018 |
toc.classList.toggle("active"); |
| 2019 |
}); |
| 2020 |
toc.querySelector("nav").addEventListener("click", e => { |
| 2021 |
toc.classList.remove("active"); |
| 2022 |
}); |
| 2023 |
</script> |
| 2024 |
</body> |
| 2025 |
</html> |