1 | <?xml version='1.0' encoding='utf-8'?>
|
2 | <rfc xmlns:xi="http://www.w3.org/2001/XInclude" version="3" category="std" consensus="true" docName="draft-ietf-dots-signal-filter-control-07" indexInclude="true" ipr="trust200902" number="9133" scripts="Common,Latin" sortRefs="true" submissionType="IETF" symRefs="true" tocDepth="4" tocInclude="true" xml:lang="en">
|
3 | <link href="https://datatracker.ietf.org/doc/draft-ietf-dots-signal-filter-control-07" rel="prev"/>
|
4 | <link href="https://dx.doi.org/10.17487/rfc9133" rel="alternate"/>
|
5 | <link href="urn:issn:2070-1721" rel="alternate"/>
|
6 | <front>
|
7 | <title abbrev="DOTS Signal Filter Control">Controlling Filtering Rules Using Distributed Denial-of-Service Open Threat Signaling (DOTS) Signal Channel</title>
|
8 | <seriesInfo name="RFC" value="9133" stream="IETF"/>
|
9 | <author fullname="Kaname Nishizuka" initials="K." surname="Nishizuka">
|
10 | <organization>NTT Communications</organization>
|
11 | <address>
|
12 | <postal>
|
13 | <street>GranPark 16F 3-4-1 Shibaura, Minato-ku</street>
|
14 | <code>108-8118</code>
|
15 | <region>Tokyo</region>
|
16 | <country>Japan</country>
|
17 | </postal>
|
18 | <email>kaname@nttv6.jp</email>
|
19 | </address>
|
20 | </author>
|
21 | <author fullname="Mohamed Boucadair" initials="M." surname="Boucadair">
|
22 | <organization>Orange</organization>
|
23 | <address>
|
24 | <postal>
|
25 | <street/>
|
26 | <city>Rennes</city>
|
27 | <code>35000</code>
|
28 | <region/>
|
29 | <country>France</country>
|
30 | </postal>
|
31 | <email>mohamed.boucadair@orange.com</email>
|
32 | </address>
|
33 | </author>
|
34 | <author fullname="Tirumaleswar Reddy.K" initials="T." surname="Reddy.K">
|
35 | <organization abbrev="Akamai">Akamai</organization>
|
36 | <address>
|
37 | <postal>
|
38 | <street>Embassy Golf Link Business Park</street>
|
39 | <city>Bangalore</city>
|
40 | <code>560071</code>
|
41 | <region>Karnataka</region>
|
42 | <country>India</country>
|
43 | </postal>
|
44 | <email>kondtir@gmail.com</email>
|
45 | </address>
|
46 | </author>
|
47 | <author fullname="Takahiko Nagata" initials="T." surname="Nagata">
|
48 | <organization>Lepidum</organization>
|
49 | <address>
|
50 | <postal>
|
51 | <street/>
|
52 | <city/>
|
53 | <region/>
|
54 | <code/>
|
55 | <country>Japan</country>
|
56 | </postal>
|
57 | <email>nagata@lepidum.co.jp</email>
|
58 | </address>
|
59 | </author>
|
60 | <date month="09" year="2021"/>
|
61 | <workgroup>DOTS</workgroup>
|
62 | <keyword>Mitigation</keyword>
|
63 | <keyword>Automation</keyword>
|
64 | <keyword>Filtering</keyword>
|
65 | <keyword>Protective Networking</keyword>
|
66 | <keyword>Protected Networks</keyword>
|
67 | <keyword>Security</keyword>
|
68 | <keyword>Anti-DDoS</keyword>
|
69 | <keyword>Reactive</keyword>
|
70 | <keyword>Collaborative Networking</keyword>
|
71 | <keyword>Collaborative Security</keyword>
|
72 | <abstract>
|
73 | <t>This document specifies an extension to the Distributed
|
74 | Denial-of-Service Open Threat Signaling (DOTS) signal channel protocol
|
75 | so that DOTS clients can control their filtering rules when an attack
|
76 | mitigation is active.</t>
|
77 | <t>Particularly, this extension allows a DOTS client to activate or
|
78 | deactivate existing filtering rules during a Distributed
|
79 | Denial-of-Service (DDoS) attack. The
|
80 | characterization of these filtering rules is conveyed by a DOTS client
|
81 | during an 'idle' time (i.e., no mitigation is active) by means of the DOTS
|
82 | data channel protocol.</t>
|
83 | </abstract>
|
84 | </front>
|
85 | <middle>
|
86 | <section anchor="introduction">
|
87 | <name>Introduction</name>
|
88 | <section anchor="problem">
|
89 | <name>The Problem</name>
|
90 | <t>In the Distributed Denial-of-Service Open Threat Signaling (DOTS)
|
91 | architecture <xref target="RFC8811"/>, DOTS
|
92 | clients and servers communicate using both a signal channel protocol
|
93 | <xref target="RFC9132"/> and a data channel protocol <xref target="RFC8783"/>.</t>
|
94 | <t>The DOTS data channel protocol <xref target="RFC8783"/> is used for bulk data exchange between DOTS agents
|
95 | to improve the coordination of parties involved in the response to a
|
96 | Distributed Denial-of-Service (DDoS) attack. Filter management, which
|
97 | is one of the tasks of the DOTS data channel protocol, enables a DOTS
|
98 | client to retrieve the filtering capabilities of a DOTS server and to
|
99 | manage filtering rules. Typically, these filtering rules are used for
|
100 | dropping or rate-limiting unwanted traffic, and permitting
|
101 | accept-listed traffic.</t>
|
102 | <t>The DOTS signal channel protocol <xref target="RFC9132"/> is
|
103 | designed to be resilient under extremely hostile network conditions
|
104 | and provides continued contact between DOTS agents even as DDoS attack
|
105 | traffic saturates the link. The DOTS signal channel can be established
|
106 | between two DOTS agents prior to or during an attack. At any time, the
|
107 | DOTS client may send mitigation requests (as per <xref target="RFC9132" section="4.4"/>) to a DOTS server over the active signal
|
108 | channel. While mitigation is active, the DOTS server periodically
|
109 | sends status messages to the DOTS client, including basic mitigation
|
110 | feedback details. In case of a massive DDoS attack that saturates the
|
111 | incoming link(s) to the DOTS client, all traffic from the DOTS server
|
112 | to the DOTS client will likely be dropped. However, the DOTS server
|
113 | may still receive DOTS messages sent from the DOTS client over the
|
114 | signaling channel thanks to the heartbeat requests keeping the
|
115 | channel active (as described in <xref target="RFC9132" section="4.7"/>).</t>
|
116 | <t>Unlike the DOTS signal channel protocol, the DOTS data channel
|
117 | protocol is not expected to deal with attack conditions. As such, an
|
118 | issue that might be encountered in some deployments is when filters
|
119 | installed by means of the DOTS data channel protocol may not function
|
120 | as expected during DDoS attacks or, worse, exacerbate an ongoing DDoS
|
121 | attack. In such conditions, the DOTS data channel protocol cannot be
|
122 | used to change these filters, which may complicate DDoS mitigation
|
123 | operations <xref target="INTEROP"/>.</t>
|
124 | <t>A typical case is a conflict between filtering rules installed by a
|
125 | DOTS client and the mitigation actions of a DDoS mitigator. Consider,
|
126 | for instance, a DOTS client that configures during 'idle' time (i.e.,
|
127 | no mitigation is active) some filtering rules using the DOTS data
|
128 | channel protocol to permit traffic from accept-listed sources.
|
129 | However, during a volumetric DDoS attack, the DDoS mitigator identifies
|
130 | the source addresses/prefixes in the accept-listed filtering rules are
|
131 | attacking the target. For example, an attacker can spoof the IP
|
132 | addresses of accept-listed sources to generate attack traffic, or the
|
133 | attacker can compromise the accept-listed sources and program them to
|
134 | launch a DDoS attack.</t>
|
135 | <t><xref target="RFC9132"/> is designed so that the
|
136 | DDoS server notifies the above conflict to the DOTS client (that is,
|
137 | the 'conflict-cause' parameter is set to 2 (conflict-with-acceptlist)),
|
138 | but the DOTS client may not be able to withdraw the accept-list rules
|
139 | during the attack period due to the high-volume attack traffic
|
140 | saturating the inbound link to the DOTS client domain. In other
|
141 | words, the DOTS client cannot use the DOTS data channel protocol to
|
142 | withdraw the accept-list filters when a DDoS attack is in
|
143 | progress.</t>
|
144 | </section>
|
145 | <section anchor="sol">
|
146 | <name>Controlling Filtering Rules Using DOTS Signal Channel</name>
|
147 | <t>This specification addresses the problems discussed in <xref target="problem"/> by adding a capability for managing filtering
|
148 | rules using the DOTS signal channel protocol, which enables a DOTS
|
149 | client to request the activation (or deactivation) of filtering rules
|
150 | during a DDoS attack. Note that creating these filtering rules is
|
151 | still the responsibility of the DOTS data channel <xref target="RFC8783"/>.</t>
|
152 | <t>The DOTS signal channel protocol is designed to enable a DOTS
|
153 | client to contact a DOTS server for help even under severe network
|
154 | congestion conditions. Therefore, extending the DOTS signal channel
|
155 | protocol to manage the filtering rules during an attack will enhance
|
156 | the protection capability offered by DOTS protocols.</t>
|
157 | <aside>
|
158 | <t>Note: The experiment at the IETF 103 hackathon <xref target="INTEROP"/> showed that even when the
|
159 | inbound link is saturated by DDoS attack traffic, the DOTS client
|
160 | can signal mitigation requests using the DOTS signal channel over
|
161 | the saturated link.</t>
|
162 | </aside>
|
163 | <t>Conflicts that are induced by filters installed by other DOTS
|
164 | clients of the same domain are not discussed in this
|
165 | specification.</t>
|
166 | <t>An augmentation to the DOTS signal channel YANG module is defined
|
167 | in <xref target="YANG"/>.</t>
|
168 | <t>Sample examples are provided in <xref target="sample"/>, in
|
169 | particular: </t>
|
170 | <ul>
|
171 | <li>
|
172 | <xref target="sample1"/> illustrates how the filter
|
173 | control extension is used when conflicts with Access Control Lists
|
174 | (ACLs) are detected and reported by a DOTS server.</li>
|
175 | <li>
|
176 | <xref target="sample2"/> shows how a DOTS client can
|
177 | instruct a DOTS server to safely forward some specific traffic in
|
178 | 'attack' time.</li>
|
179 | <li>
|
180 | <xref target="sample3"/> shows how a DOTS client can
|
181 | react if the DDoS traffic is still being forwarded to the DOTS
|
182 | client domain even if mitigation requests were sent to a DOTS
|
183 | server.</li>
|
184 | </ul>
|
185 | <t>The JavaScript Object Notation (JSON) encoding of YANG-modeled data
|
186 | <xref target="RFC7951"/> is used to illustrate the examples.</t>
|
187 | </section>
|
188 | </section>
|
189 | <section anchor="notation">
|
190 | <name>Terminology</name>
|
191 | <t>The key words "<bcp14>MUST</bcp14>", "<bcp14>MUST NOT</bcp14>",
|
192 | "<bcp14>REQUIRED</bcp14>", "<bcp14>SHALL</bcp14>", "<bcp14>SHALL NOT</bcp14>", "<bcp14>SHOULD</bcp14>", "<bcp14>SHOULD NOT</bcp14>",
|
193 | "<bcp14>RECOMMENDED</bcp14>", "<bcp14>NOT RECOMMENDED</bcp14>",
|
194 | "<bcp14>MAY</bcp14>", and "<bcp14>OPTIONAL</bcp14>" in this document are
|
195 | to be interpreted as described in BCP 14 <xref target="RFC2119"/> <xref target="RFC8174"/> when, and
|
196 | only when, they appear in all capitals, as shown here.</t>
|
197 | <t>The reader should be familiar with the terms defined in <xref target="RFC8612"/>.</t>
|
198 | <t>The terminology for describing YANG modules is defined in <xref target="RFC7950"/>. The meaning of the symbols in the
|
199 | tree diagram is defined in <xref target="RFC8340"/> and <xref target="RFC8791"/>.</t>
|
200 | </section>
|
201 | <section>
|
202 | <name>Controlling Filtering Rules of a DOTS Client</name>
|
203 | <section anchor="bind">
|
204 | <name>Binding DOTS Data and Signal Channels</name>
|
205 | <t>The filtering rules eventually managed using the DOTS signal
|
206 | channel protocol are created a priori by the same DOTS client using
|
207 | the DOTS data channel protocol. Managing conflicts with filters
|
208 | installed by other DOTS clients of the same domain is out of
|
209 | scope.</t>
|
210 | <t>As discussed in <xref target="RFC9132" section="4.4.1"/>, a DOTS client must use the same 'cuid' for both the
|
211 | DOTS signal and data channels. This requirement is meant to facilitate
|
212 | binding DOTS channels used by the same DOTS client.</t>
|
213 | <t>The DOTS signal and data channels from a DOTS client may or may not
|
214 | use the same DOTS server. Nevertheless, the scope of the mitigation
|
215 | request, alias, and filtering rules are not restricted to the DOTS
|
216 | server but to the DOTS server domain. To that aim, DOTS servers within
|
217 | a domain are assumed to have a mechanism to coordinate the mitigation
|
218 | requests, aliases, and filtering rules to coordinate their decisions
|
219 | for better mitigation operation efficiency. The exact details about
|
220 | such a mechanism is out of the scope of this document.</t>
|
221 | <t>A filtering rule controlled by the DOTS signal channel is
|
222 | identified by its ACL name (<xref target="RFC8783" section="4.3"/>). Note that an ACL name unambiguously
|
223 | identifies an ACL bound to a DOTS client, but the same name may be
|
224 | used by distinct DOTS clients.</t>
|
225 | <t>The activation or deactivation of an ACL by the DOTS signal channel
|
226 | overrides the 'activation-type' (defined in <xref target="RFC8783" section="4.3"/>) a priori conveyed with the
|
227 | filtering rules using the DOTS data channel protocol.</t>
|
228 | <t>Once the attack is mitigated, the DOTS client may use the data
|
229 | channel to control the 'activation-type' (e.g., revert to a default
|
230 | value) of some of the filtering rules controlled by the DOTS signal
|
231 | channel or delete some of these filters. This behavior is deployment
|
232 | specific.</t>
|
233 | </section>
|
234 | <section>
|
235 | <name>DOTS Signal Channel Extension</name>
|
236 | <section anchor="filtering">
|
237 | <name>Parameters and Behaviors</name>
|
238 | <t>This specification extends the mitigation request defined in
|
239 | <xref target="RFC9132" section="4.4.1"/> to
|
240 | convey the intended control of configured filtering
|
241 | rules. Concretely, the DOTS client conveys the 'acl-list' attribute with
|
242 | the following sub-attributes in the Concise Binary Object
|
243 | Representation (CBOR) body of a mitigation
|
244 | request (see the YANG structure in <xref target="tree"/>):</t>
|
245 | <dl>
|
246 | <dt>acl-name:</dt>
|
247 | <dd>
|
248 | <t>A name of an access list defined using
|
249 | the DOTS data channel (<xref target="RFC8783" section="4.3"/>) that is associated with the DOTS
|
250 | client.</t>
|
251 | <t>As a reminder, an ACL is an ordered list of Access Control
|
252 | Entries (ACEs). Each ACE has a list of match criteria and a list
|
253 | of actions <xref target="RFC8783"/>. The list
|
254 | of configured ACLs can be retrieved using the DOTS data channel
|
255 | during 'idle' time.</t>
|
256 | <t>This is a mandatory attribute when 'acl-list'
|
257 | is included.</t>
|
258 | </dd>
|
259 | <dt>activation-type:</dt>
|
260 | <dd>
|
261 | <t>An attribute indicating the activation type of
|
262 | an ACL overriding the existing 'activation-type' installed by
|
263 | the DOTS client using the DOTS data channel. </t>
|
264 | <t>As a reminder, this attribute can be set to
|
265 | 'deactivate', 'immediate', or 'activate-when-mitigating' as
|
266 | defined in <xref target="RFC8783"/>. </t>
|
267 | <t>Note that both 'immediate' and
|
268 | 'activate-when-mitigating' have an immediate effect when a
|
269 | mitigation request is being processed by the DOTS server.
|
270 | </t>
|
271 | <t>This is an optional attribute.</t>
|
272 | </dd>
|
273 | </dl>
|
274 | <t>By default, ACL-related operations are achieved using the DOTS
|
275 | data channel protocol when no attack is ongoing. DOTS clients <bcp14>MUST NOT</bcp14> use the filtering control over the DOTS signal channel in 'idle'
|
276 | time; such requests <bcp14>MUST</bcp14> be discarded by DOTS servers with 4.00 (Bad
|
277 | Request).</t>
|
278 | <t>During an attack time, DOTS clients may include 'acl-list',
|
279 | 'acl-name', and 'activation-type' attributes in a mitigation
|
280 | request. This request may be the initial mitigation request for a
|
281 | given mitigation scope or a new one overriding an existing request.
|
282 | In both cases, a new 'mid' <bcp14>MUST</bcp14> be used. Nevertheless, it is <bcp14>NOT RECOMMENDED</bcp14> to include ACL attributes in an initial mitigation
|
283 | request for a given mitigation scope or in a mitigation request
|
284 | adjusting the mitigation scope. This recommendation is meant to
|
285 | avoid delaying attack mitigations because of failures to process ACL
|
286 | attributes.</t>
|
287 | <t>As the attack evolves, DOTS clients can adjust the
|
288 | 'activation-type' of an ACL conveyed in a mitigation request or
|
289 | control other filters as necessary. This can be achieved by sending
|
290 | a PUT request with a new 'mid' value.</t>
|
291 | <t>It is <bcp14>RECOMMENDED</bcp14> for a DOTS client to subscribe
|
292 | to asynchronous notifications of the attack mitigation, as detailed
|
293 | in <xref target="RFC9132" section="4.4.2.1"/>. If
|
294 | not, the polling mechanism in <xref target="RFC9132" section="4.4.2.2"/> has to be followed by the
|
295 | DOTS client.</t>
|
296 | <t>A DOTS client relies on the information received from the DOTS
|
297 | server and/or local information to the DOTS client domain to trigger
|
298 | a filter control request. Only filters that are pertinent for an
|
299 | ongoing mitigation should be controlled by a DOTS client using the
|
300 | DOTS signal channel.</t>
|
301 | <t>'acl-list', 'acl-name', and 'activation-type' are defined as
|
302 | comprehension-required parameters. Following the rules in <xref target="RFC9132" section="6"/>, if the DOTS
|
303 | server does not understand the 'acl-list', 'acl-name', or
|
304 | 'activation-type' attributes, it responds with a 4.00 (Bad
|
305 | Request) error response code.</t>
|
306 | <t>If the DOTS server does not find the ACL name ('acl-name')
|
307 | conveyed in the mitigation request for this DOTS client, it <bcp14>MUST</bcp14>
|
308 | respond with a 4.04 (Not Found) error response code.</t>
|
309 | <t>If the DOTS server finds the ACL name for this DOTS client, and
|
310 | assuming the request passed the validation checks in <xref target="RFC9132" section="4.4.1"/>, the DOTS
|
311 | server <bcp14>MUST</bcp14> proceed with the 'activation-type'
|
312 | update. The update is immediately enforced by the DOTS server and
|
313 | will be maintained as the new activation type for the ACL name even
|
314 | after the termination of the mitigation request. In addition, the
|
315 | DOTS server <bcp14>MUST</bcp14> update the lifetime of the
|
316 | corresponding ACL similar to the update when a refresh request is
|
317 | received using the DOTS data channel (<xref target="RFC8783" section="7.2"/>). If, for some reason, the DOTS
|
318 | server fails to apply the filter update, it <bcp14>MUST</bcp14>
|
319 | respond with a 5.03 (Service Unavailable) error response code and
|
320 | include the failed ACL update in the diagnostic payload of the
|
321 | response (an example is shown in <xref target="diag"/>). Else, the DOTS server replies with the
|
322 | appropriate response code defined in <xref target="RFC9132" section="4.4.1"/>.</t>
|
323 | <figure anchor="diag">
|
324 | <name>Example of a Diagnostic Payload Including Failed ACL Update</name>
|
325 | <sourcecode>
|
326 | {
|
327 | "ietf-dots-signal-channel:mitigation-scope": {
|
328 | "scope": [
|
329 | {
|
330 | "mid": 123,
|
331 | "ietf-dots-signal-control:acl-list": [
|
332 | {
|
333 | "acl-name": "an-accept-list",
|
334 | "activation-type": "deactivate"
|
335 | }
|
336 | ]
|
337 | }
|
338 | ]
|
339 | }
|
340 | }
|
341 | </sourcecode>
|
342 | </figure>
|
343 | <t>The JSON/YANG mappings for DOTS filter control attributes are
|
344 | shown in <xref target="table1"/>. As a reminder, the mapping for 'acl-name' is
|
345 | defined in Table 5 of <xref target="RFC9132"/>.</t>
|
346 | <table anchor="table1">
|
347 | <name>JSON/YANG Mapping to CBOR for Filter Control Attributes</name>
|
348 | <thead>
|
349 | <tr>
|
350 | <th>Parameter Name</th>
|
351 | <th>YANG Type</th>
|
352 | <th>CBOR Key</th>
|
353 | <th>CBOR Major Type & Information</th>
|
354 | <th>JSON Type</th>
|
355 | </tr>
|
356 | </thead>
|
357 | <tbody>
|
358 | <tr>
|
359 | <td>activation-type</td>
|
360 | <td>enumeration</td>
|
361 | <td>52</td>
|
362 | <td>0 unsigned</td>
|
363 | <td>String</td>
|
364 | </tr>
|
365 | <tr>
|
366 | <td>ietf-dots-signal-control:acl-list</td>
|
367 | <td>list</td>
|
368 | <td>53</td>
|
369 | <td>4 array</td>
|
370 | <td>Array</td>
|
371 | </tr>
|
372 | <tr>
|
373 | <td>acl-name</td>
|
374 | <td>leafref</td>
|
375 | <td>23</td>
|
376 | <td>3 text string</td>
|
377 | <td>String</td>
|
378 | </tr>
|
379 | </tbody>
|
380 | </table>
|
381 | <t>If the DOTS client receives a 5.03 (Service Unavailable) with a
|
382 | diagnostic payload indicating a failed ACL update as a response to
|
383 | an initial mitigation or a mitigation with adjusted scope, the DOTS
|
384 | client <bcp14>MUST</bcp14> immediately send a new request that
|
385 | repeats all the parameters as sent in the failed mitigation request
|
386 | but without including the ACL attributes. After the expiry of
|
387 | Max-Age returned in the 5.03 (Service Unavailable) response, the
|
388 | DOTS client retries with a new mitigation request (i.e., a new
|
389 | 'mid') that repeats all the parameters as sent in the failed
|
390 | mitigation request (i.e., the one including the ACL attributes).</t>
|
391 | <t>If, during an active mitigation, the 'activation-type' is changed
|
392 | at the DOTS server (e.g., as a result of an external action) for an
|
393 | ACL bound to a DOTS client, the DOTS server notifies that DOTS
|
394 | client of the change by including the corresponding ACL parameters
|
395 | in an asynchronous notification (the DOTS client is observing the
|
396 | active mitigation) or in a response to a polling request (<xref target="RFC9132" section="4.4.2.2"/>).</t>
|
397 | <t>If the DOTS signal and data channels of a DOTS client are not
|
398 | established with the same DOTS server of a DOTS server domain, the
|
399 | above request processing operations are undertaken using the
|
400 | coordination mechanism discussed in <xref target="bind"/>.</t>
|
401 | <t>This specification does not require any modification to the
|
402 | efficacy update and the withdrawal procedures defined in <xref target="RFC9132"/>. In particular, ACL-related clauses are not
|
403 | included in a PUT request used to send an efficacy update and DELETE
|
404 | requests.</t>
|
405 | </section>
|
406 | <section anchor="YANG">
|
407 | <name>DOTS Signal Filtering Control Module</name>
|
408 | <section anchor="tree">
|
409 | <name>Tree Structure</name>
|
410 | <t>This document augments the "ietf-dots-signal-channel" YANG
|
411 | module defined in <xref target="RFC9132"/> for managing
|
412 | filtering rules.</t>
|
413 | <t>This document defines the YANG module
|
414 | "ietf-dots-signal-control", which has the following tree
|
415 | structure:</t>
|
416 | <sourcecode type="yangtree">
|
417 | module: ietf-dots-signal-control
|
418 | augment-structure /dots-signal:dots-signal/dots-signal:message-type
|
419 | /dots-signal:mitigation-scope/dots-signal:scope:
|
420 | +-- acl-list* [acl-name]
|
421 | +-- acl-name
|
422 | | -> /data-channel:dots-data/dots-client/acls/acl/name
|
423 | +-- activation-type? data-channel:activation-type
|
424 | </sourcecode>
|
425 | </section>
|
426 | <section>
|
427 | <name>YANG Module</name>
|
428 | <t>This YANG module is not intended to be used via
|
429 | NETCONF/RESTCONF for DOTS server management purposes; such a module
|
430 | is out of the scope of this document. It serves only to provide a
|
431 | data model and encoding, but not a management data model.</t>
|
432 | <t>This module uses types defined in <xref target="RFC8783"/>.</t>
|
433 | <sourcecode name="ietf-dots-signal-control@2021-09-02.yang" type="yang" markers="true">
|
434 | module ietf-dots-signal-control {
|
435 | yang-version 1.1;
|
436 | namespace "urn:ietf:params:xml:ns:yang:ietf-dots-signal-control";
|
437 | prefix dots-control;
|
438 |
|
439 | import ietf-dots-signal-channel {
|
440 | prefix dots-signal;
|
441 | reference
|
442 | "RFC 9132: Distributed Denial-of-Service Open Threat
|
443 | Signaling (DOTS) Signal Channel Specification";
|
444 | }
|
445 |
|
446 |
|
447 | import ietf-yang-structure-ext {
|
448 | prefix sx;
|
449 | reference
|
450 | "RFC 8791: YANG Data Structure Extensions";
|
451 | }
|
452 |
|
453 | import ietf-dots-data-channel {
|
454 | prefix data-channel;
|
455 | reference
|
456 | "RFC 8783: Distributed Denial-of-Service Open Threat
|
457 | Signaling (DOTS) Data Channel Specification";
|
458 | }
|
459 |
|
460 | organization
|
461 | "IETF DDoS Open Threat Signaling (DOTS) Working Group";
|
462 | contact
|
463 | "WG Web: <https://datatracker.ietf.org/wg/dots/>
|
464 | WG List: <mailto:dots@ietf.org>
|
465 |
|
466 | Author: Kaname Nishizuka
|
467 | <mailto:kaname@nttv6.jp>
|
468 |
|
469 | Author: Mohamed Boucadair
|
470 | <mailto:mohamed.boucadair@orange.com>
|
471 |
|
472 | Author: Tirumaleswar Reddy.K
|
473 | <mailto:kondtir@gmail.com>
|
474 |
|
475 | Author: Takahiko Nagata
|
476 | <mailto:nagata@lepidum.co.jp>";
|
477 |
|
478 | description
|
479 | "This module contains YANG definition for the signaling
|
480 | messages exchanged between a DOTS client and a DOTS server
|
481 | to control, by means of the DOTS signal channel, filtering
|
482 | rules configured using the DOTS data channel.
|
483 |
|
484 | Copyright (c) 2021 IETF Trust and the persons identified as
|
485 | authors of the code. All rights reserved.
|
486 |
|
487 | Redistribution and use in source and binary forms, with or
|
488 | without modification, is permitted pursuant to, and subject
|
489 | to the license terms contained in, the Simplified BSD License
|
490 | set forth in Section 4.c of the IETF Trust's Legal Provisions
|
491 | Relating to IETF Documents
|
492 | (https://trustee.ietf.org/license-info).
|
493 |
|
494 | This version of this YANG module is part of RFC 9133; see
|
495 | the RFC itself for full legal notices.";
|
496 |
|
497 | revision 2021-09-02 {
|
498 | description
|
499 | "Initial revision.";
|
500 | reference
|
501 | "RFC 9133: Controlling Filtering Rules Using Distributed
|
502 | Denial-of-Service Open Threat Signaling (DOTS)
|
503 | Signal Channel";
|
504 | }
|
505 |
|
506 | sx:augment-structure "/dots-signal:dots-signal"
|
507 | + "/dots-signal:message-type"
|
508 | + "/dots-signal:mitigation-scope"
|
509 | + "/dots-signal:scope" {
|
510 |
|
511 | description
|
512 | "ACL name and activation type.";
|
513 |
|
514 | list acl-list {
|
515 | key "acl-name";
|
516 | description
|
517 | "List of ACLs as defined using the DOTS data
|
518 | channel. ACLs bound to a DOTS client are uniquely
|
519 | identified by a name.";
|
520 | leaf acl-name {
|
521 | type leafref {
|
522 | path "/data-channel:dots-data/data-channel:dots-client"
|
523 | + "/data-channel:acls/data-channel:acl"
|
524 | + "/data-channel:name";
|
525 | }
|
526 | description
|
527 | "Reference to the ACL name bound to a DOTS client.";
|
528 | }
|
529 | leaf activation-type {
|
530 | type data-channel:activation-type;
|
531 | default "activate-when-mitigating";
|
532 | description
|
533 | "Sets the activation type of an ACL.";
|
534 | }
|
535 | }
|
536 | }
|
537 | }
|
538 | </sourcecode>
|
539 | </section>
|
540 | </section>
|
541 | </section>
|
542 | </section>
|
543 | <section anchor="sample">
|
544 | <name>Some Examples</name>
|
545 | <t>This section provides some examples to illustrate the behavior
|
546 | specified in <xref target="filtering"/>. These examples are
|
547 | provided for illustration purposes; they should not be considered as
|
548 | deployment recommendations.</t>
|
549 | <section anchor="sample1">
|
550 | <name>Conflict Handling</name>
|
551 | <t>Let's consider a DOTS client that contacts its DOTS server during
|
552 | 'idle' time to install an accept-list allowing for UDP traffic issued
|
553 | from 2001:db8:1234::/48 with a destination port number 443 to be
|
554 | forwarded to 2001:db8:6401::2/127. It does so by sending, for example,
|
555 | a PUT request as shown in <xref target="PUT"/>.</t>
|
556 | <figure anchor="PUT">
|
557 | <name>DOTS Data Channel Request to Create a Filter</name>
|
558 | <sourcecode>
|
559 | PUT /restconf/data/ietf-dots-data-channel:dots-data\
|
560 | /dots-client=paL8p4Zqo4SLv64TLPXrxA/acls\
|
561 | /acl=an-accept-list HTTP/1.1
|
562 | Host: example.com
|
563 | Content-Type: application/yang-data+json
|
564 |
|
565 | {
|
566 | "ietf-dots-data-channel:acls": {
|
567 | "acl": [
|
568 | {
|
569 | "name": "an-accept-list",
|
570 | "type": "ipv6-acl-type",
|
571 | "activation-type": "activate-when-mitigating",
|
572 | "aces": {
|
573 | "ace": [
|
574 | {
|
575 | "name": "test-ace-ipv6-udp",
|
576 | "matches": {
|
577 | "ipv6": {
|
578 | "destination-ipv6-network": "2001:db8:6401::2/127",
|
579 | "source-ipv6-network": "2001:db8:1234::/48"
|
580 | },
|
581 | "udp": {
|
582 | "destination-port-range-or-operator": {
|
583 | "operator": "eq",
|
584 | "port": 443
|
585 | }
|
586 | }
|
587 | },
|
588 | "actions": {
|
589 | "forwarding": "accept"
|
590 | }
|
591 | }
|
592 | ]
|
593 | }
|
594 | }
|
595 | ]
|
596 | }
|
597 | }
|
598 | </sourcecode>
|
599 | </figure>
|
600 | <t>Sometime later, consider that a DDoS attack is detected by the DOTS
|
601 | client on 2001:db8:6401::2/127. Consequently, the DOTS client sends a
|
602 | mitigation request to its DOTS server as shown in <xref target="mitigate"/>.</t>
|
603 | <figure anchor="mitigate">
|
604 | <name>DOTS Signal Channel Mitigation Request</name>
|
605 | <sourcecode>
|
606 | Header: PUT (Code=0.03)
|
607 | Uri-Path: ".well-known"
|
608 | Uri-Path: "dots"
|
609 | Uri-Path: "mitigate"
|
610 | Uri-Path: "cuid=paL8p4Zqo4SLv64TLPXrxA"
|
611 | Uri-Path: "mid=123"
|
612 | Content-Format: "application/dots+cbor"
|
613 |
|
614 | {
|
615 | "ietf-dots-signal-channel:mitigation-scope": {
|
616 | "scope": [
|
617 | {
|
618 | "target-prefix": [
|
619 | "2001:db8:6401::2/127"
|
620 | ],
|
621 | "target-protocol": [
|
622 | 17
|
623 | ],
|
624 | "lifetime": 3600
|
625 | }
|
626 | ]
|
627 | }
|
628 | }
|
629 | </sourcecode>
|
630 | </figure>
|
631 | <t>The DOTS server immediately accepts the request by replying with
|
632 | 2.01 (Created) (<xref target="response"/> depicts the message
|
633 | body of the response).</t>
|
634 | <figure anchor="response">
|
635 | <name>Status Response (Message Body)</name>
|
636 | <sourcecode>
|
637 | {
|
638 | "ietf-dots-signal-channel:mitigation-scope": {
|
639 | "scope": [
|
640 | {
|
641 | "mid": 123,
|
642 | "lifetime": 3600
|
643 | }
|
644 | ]
|
645 | }
|
646 | }
|
647 | </sourcecode>
|
648 | </figure>
|
649 | <t>Assuming the DOTS client subscribed to asynchronous notifications,
|
650 | when the DOTS server concludes that some of the attack sources belong
|
651 | to 2001:db8:1234::/48, it sends a notification message with 'status'
|
652 | code set to 1 (attack-mitigation-in-progress) and 'conflict-cause' set
|
653 | to 2 (conflict-with-acceptlist) to the DOTS client to indicate that
|
654 | this mitigation request is in progress, but a conflict is
|
655 | detected.</t>
|
656 | <t>Upon receipt of the notification message from the DOTS server, the
|
657 | DOTS client sends a PUT request to deactivate the "an-accept-list" ACL
|
658 | as shown in <xref target="control"/>.</t>
|
659 | <t>The DOTS client can also decide to send a PUT request to deactivate
|
660 | the "an-accept-list" ACL if suspect traffic is received from an
|
661 | accept-listed source (2001:db8:1234::/48). The structure of that PUT
|
662 | is the same as the one shown in <xref target="control"/>.</t>
|
663 | <figure anchor="control">
|
664 | <name>PUT for Deactivating a Conflicting Filter</name>
|
665 | <sourcecode>
|
666 | Header: PUT (Code=0.03)
|
667 | Uri-Path: ".well-known"
|
668 | Uri-Path: "dots"
|
669 | Uri-Path: "mitigate"
|
670 | Uri-Path: "cuid=paL8p4Zqo4SLv64TLPXrxA"
|
671 | Uri-Path: "mid=124"
|
672 | Content-Format: "application/dots+cbor"
|
673 |
|
674 | {
|
675 | "ietf-dots-signal-channel:mitigation-scope": {
|
676 | "scope": [
|
677 | {
|
678 | "target-prefix": [
|
679 | "2001:db8:6401::2/127"
|
680 | ],
|
681 | "target-protocol": [
|
682 | 17
|
683 | ],
|
684 | "ietf-dots-signal-control:acl-list": [
|
685 | {
|
686 | "acl-name": "an-accept-list",
|
687 | "activation-type": "deactivate"
|
688 | }
|
689 | ],
|
690 | "lifetime": 3600
|
691 | }
|
692 | ]
|
693 | }
|
694 | }
|
695 | </sourcecode>
|
696 | </figure>
|
697 | <t>Then, the DOTS server deactivates the "an-accept-list" ACL and replies
|
698 | with a 2.04 (Changed) response to the DOTS client to confirm the
|
699 | successful operation. The message body is similar to the one depicted
|
700 | in <xref target="response"/>.</t>
|
701 | <t>Once the attack is mitigated, the DOTS client may use the data
|
702 | channel to retrieve its ACLs maintained by the DOTS server. As shown
|
703 | in <xref target="GET-2"/>, the activation type is set to
|
704 | 'deactivate' as set by the DOTS signal channel (<xref target="control"/>) instead of the type initially set using the
|
705 | DOTS data channel (<xref target="PUT"/>).</t>
|
706 | <figure anchor="GET-2">
|
707 | <name>DOTS Data Channel GET Response after Mitigation (Message Body)</name>
|
708 | <sourcecode>
|
709 | {
|
710 | "ietf-dots-data-channel:acls": {
|
711 | "acl": [
|
712 | {
|
713 | "name": "an-accept-list",
|
714 | "type": "ipv6-acl-type",
|
715 | "activation-type": "deactivate",
|
716 | "pending-lifetime": 10021,
|
717 | "aces": {
|
718 | "ace": [
|
719 | {
|
720 | "name": "test-ace-ipv6-udp",
|
721 | "matches": {
|
722 | "ipv6": {
|
723 | "destination-ipv6-network": "2001:db8:6401::2/127",
|
724 | "source-ipv6-network": "2001:db8:1234::/48"
|
725 | },
|
726 | "udp": {
|
727 | "destination-port-range-or-operator": {
|
728 | "operator": "eq",
|
729 | "port": 443
|
730 | }
|
731 | }
|
732 | },
|
733 | "actions": {
|
734 | "forwarding": "accept"
|
735 | }
|
736 | }
|
737 | ]
|
738 | }
|
739 | }
|
740 | ]
|
741 | }
|
742 | }
|
743 | </sourcecode>
|
744 | </figure>
|
745 | </section>
|
746 | <section anchor="sample2">
|
747 | <name>On-Demand Activation of an Accept-List Filter</name>
|
748 | <t>Let's consider a DOTS client that contacts its DOTS server during
|
749 | 'idle' time to install an accept-list allowing for UDP traffic issued
|
750 | from 2001:db8:1234::/48 to be forwarded to 2001:db8:6401::2/127. It
|
751 | does so by sending, for example, a PUT request shown in <xref target="PUT1"/>. The DOTS server installs this filter
|
752 | with a "deactivated" state.</t>
|
753 | <figure anchor="PUT1">
|
754 | <name>DOTS Data Channel Request to Create an Accept-List Filter</name>
|
755 | <sourcecode>
|
756 | PUT /restconf/data/ietf-dots-data-channel:dots-data\
|
757 | /dots-client=ioiuLoZqo4SLv64TLPXrxA/acls\
|
758 | /acl=my-accept-list HTTP/1.1
|
759 | Host: example.com
|
760 | Content-Type: application/yang-data+json
|
761 |
|
762 | {
|
763 | "ietf-dots-data-channel:acls": {
|
764 | "acl": [
|
765 | {
|
766 | "name": "my-accept-list",
|
767 | "type": "ipv6-acl-type",
|
768 | "activation-type": "deactivate",
|
769 | "aces": {
|
770 | "ace": [
|
771 | {
|
772 | "name": "an-ace",
|
773 | "matches": {
|
774 | "ipv6": {
|
775 | "destination-ipv6-network": "2001:db8:6401::2/127",
|
776 | "source-ipv6-network": "2001:db8:1234::/48",
|
777 | "protocol": 17
|
778 | }
|
779 | },
|
780 | "actions": {
|
781 | "forwarding": "accept"
|
782 | }
|
783 | }
|
784 | ]
|
785 | }
|
786 | }
|
787 | ]
|
788 | }
|
789 | }
|
790 | </sourcecode>
|
791 | </figure>
|
792 | <t>Sometime later, consider that a UDP DDoS attack is detected by the
|
793 | DOTS client on 2001:db8:6401::2/127 but the DOTS client wants to let
|
794 | the traffic from 2001:db8:1234::/48 be accept-listed to the DOTS
|
795 | client domain. Consequently, the DOTS client sends a mitigation
|
796 | request to its DOTS server as shown in <xref target="mitigate1"/>.</t>
|
797 | <figure anchor="mitigate1">
|
798 | <name>DOTS Signal Channel Mitigation Request with a Filter Control</name>
|
799 | <sourcecode>
|
800 | Header: PUT (Code=0.03)
|
801 | Uri-Path: ".well-known"
|
802 | Uri-Path: "dots"
|
803 | Uri-Path: "mitigate"
|
804 | Uri-Path: "cuid=ioiuLoZqo4SLv64TLPXrxA"
|
805 | Uri-Path: "mid=4879"
|
806 | Content-Format: "application/dots+cbor"
|
807 |
|
808 | {
|
809 | "ietf-dots-signal-channel:mitigation-scope": {
|
810 | "scope": [
|
811 | {
|
812 | "target-prefix": [
|
813 | "2001:db8:6401::2/127"
|
814 | ],
|
815 | "target-protocol": [
|
816 | 17
|
817 | ],
|
818 | "ietf-dots-signal-control:acl-list": [
|
819 | {
|
820 | "acl-name": "my-accept-list",
|
821 | "activation-type": "immediate"
|
822 | }
|
823 | ],
|
824 | "lifetime": 3600
|
825 | }
|
826 | ]
|
827 | }
|
828 | }
|
829 | </sourcecode>
|
830 | </figure>
|
831 | <t>The DOTS server activates the "my-accept-list" ACL and replies with
|
832 | a 2.01 (Created) response to the DOTS client to confirm the successful
|
833 | operation.</t>
|
834 | </section>
|
835 | <section anchor="sample3">
|
836 | <name>DOTS Servers/Mitigators Lacking Capacity</name>
|
837 | <t>This section describes a scenario in which a DOTS client activates
|
838 | a drop-list or a rate-limit filter during an attack.</t>
|
839 | <t>Consider a DOTS client that contacts its DOTS server during 'idle'
|
840 | time to install an accept-list that rate-limits all (or a part
|
841 | thereof) traffic to be forwarded to 2001:db8:123::/48 as a last resort
|
842 | countermeasure whenever required. Installing the accept-list can be
|
843 | done by sending, for example, the PUT request shown in <xref target="rate"/>. The DOTS server installs this filter
|
844 | with a "deactivated" state.</t>
|
845 | <figure anchor="rate">
|
846 | <name>DOTS Data Channel Request to Create a Rate-Limit Filter</name>
|
847 | <sourcecode>
|
848 | PUT /restconf/data/ietf-dots-data-channel:dots-data\
|
849 | /dots-client=OopPisZqo4SLv64TLPXrxA/acls\
|
850 | /acl=my-ratelimit-list HTTP/1.1
|
851 | Host: example.com
|
852 | Content-Type: application/yang-data+json
|
853 |
|
854 | {
|
855 | "ietf-dots-data-channel:acls": {
|
856 | "acl": [
|
857 | {
|
858 | "name": "my-ratelimit-list",
|
859 | "type": "ipv6-acl-type",
|
860 | "activation-type": "deactivate",
|
861 | "aces": {
|
862 | "ace": [
|
863 | {
|
864 | "name": "my-ace",
|
865 | "matches": {
|
866 | "ipv6": {
|
867 | "destination-ipv6-network": "2001:db8:123::/48"
|
868 | }
|
869 | },
|
870 | "actions": {
|
871 | "forwarding": "accept",
|
872 | "rate-limit": "20000.00"
|
873 | }
|
874 | }
|
875 | ]
|
876 | }
|
877 | }
|
878 | ]
|
879 | }
|
880 | }
|
881 | </sourcecode>
|
882 | </figure>
|
883 | <t>Consider now that a DDoS attack is detected by the DOTS client on
|
884 | 2001:db8:123::/48. Consequently, the DOTS client sends a mitigation
|
885 | request to its DOTS server (<xref target="ratel"/>).</t>
|
886 | <figure anchor="ratel">
|
887 | <name>DOTS Signal Channel Mitigation Request</name>
|
888 | <sourcecode>
|
889 | Header: PUT (Code=0.03)
|
890 | Uri-Path: ".well-known"
|
891 | Uri-Path: "dots"
|
892 | Uri-Path: "mitigate"
|
893 | Uri-Path: "cuid=OopPisZqo4SLv64TLPXrxA"
|
894 | Uri-Path: "mid=85"
|
895 | Content-Format: "application/dots+cbor"
|
896 |
|
897 | {
|
898 | "ietf-dots-signal-channel:mitigation-scope": {
|
899 | "scope": [
|
900 | {
|
901 | "target-prefix": [
|
902 | "2001:db8:123::/48"
|
903 | ],
|
904 | "lifetime": 3600
|
905 | }
|
906 | ]
|
907 | }
|
908 | }
|
909 | </sourcecode>
|
910 | </figure>
|
911 | <t>For some reason (e.g., the DOTS server, or the mitigator, is
|
912 | lacking a capability or capacity), the DOTS client is still receiving
|
913 | attack traffic, which saturates available links. To soften the
|
914 | problem, the DOTS client decides to activate the filter that
|
915 | rate-limits the traffic destined to the DOTS client domain. To that
|
916 | aim, the DOTS client sends the mitigation request to its DOTS server
|
917 | shown in <xref target="rateres"/>.</t>
|
918 | <figure anchor="rateres">
|
919 | <name>DOTS Signal Channel Mitigation Request to Activate a Rate-Limit Filter</name>
|
920 | <sourcecode>
|
921 | Header: PUT (Code=0.03)
|
922 | Uri-Path: ".well-known"
|
923 | Uri-Path: "dots"
|
924 | Uri-Path: "mitigate"
|
925 | Uri-Path: "cuid=OopPisZqo4SLv64TLPXrxA"
|
926 | Uri-Path: "mid=86"
|
927 | Content-Format: "application/dots+cbor"
|
928 |
|
929 | {
|
930 | "ietf-dots-signal-channel:mitigation-scope": {
|
931 | "scope": [
|
932 | {
|
933 | "target-prefix": [
|
934 | "2001:db8:123::/48"
|
935 | ],
|
936 | "ietf-dots-signal-control:acl-list": [
|
937 | {
|
938 | "acl-name": "my-ratelimit-list",
|
939 | "activation-type": "immediate"
|
940 | }
|
941 | ],
|
942 | "lifetime": 3600
|
943 | }
|
944 | ]
|
945 | }
|
946 | }
|
947 | </sourcecode>
|
948 | </figure>
|
949 | <t>Then, the DOTS server activates the "my-ratelimit-list" ACL and replies
|
950 | with a 2.04 (Changed) response to the DOTS client to confirm the
|
951 | successful operation.</t>
|
952 | <t>As the attack mitigation evolves, the DOTS client may decide to
|
953 | deactivate the rate-limit policy (e.g., upon receipt of a notification
|
954 | status change from 'attack-exceeded-capability' to
|
955 | 'attack-mitigation-in-progress'). Based on the mitigation status
|
956 | conveyed by the DOTS server, the DOTS client can deactivate the
|
957 | rate-limit action. It does so by sending the request shown in <xref target="rateres2"/>.</t>
|
958 | <figure anchor="rateres2">
|
959 | <name>DOTS Signal Channel Mitigation Request to Deactivate a Rate-Limit Filter</name>
|
960 | <sourcecode type="cbor">
|
961 | Header: PUT (Code=0.03)
|
962 | Uri-Path: ".well-known"
|
963 | Uri-Path: "dots"
|
964 | Uri-Path: "mitigate"
|
965 | Uri-Path: "cuid=OopPisZqo4SLv64TLPXrxA"
|
966 | Uri-Path: "mid=87"
|
967 | Content-Format: "application/dots+cbor"
|
968 |
|
969 | {
|
970 | "ietf-dots-signal-channel:mitigation-scope": {
|
971 | "scope": [
|
972 | {
|
973 | "target-prefix": [
|
974 | "2001:db8:123::/48"
|
975 | ],
|
976 | "ietf-dots-signal-control:acl-list": [
|
977 | {
|
978 | "acl-name": "my-ratelimit-list",
|
979 | "activation-type": "deactivate"
|
980 | }
|
981 | ],
|
982 | "lifetime": 3600
|
983 | }
|
984 | ]
|
985 | }
|
986 | }
|
987 | </sourcecode>
|
988 | </figure>
|
989 | </section>
|
990 | </section>
|
991 | <section anchor="IANA">
|
992 | <name>IANA Considerations</name>
|
993 | <section anchor="map">
|
994 | <name>DOTS Signal Channel CBOR Key Values Subregistry</name>
|
995 | <t>Per this specification, IANA has registered the following parameters in the
|
996 | "DOTS Signal Channel CBOR Key Values" subregistry within the
|
997 | "Distributed Denial-of-Service Open Threat Signaling (DOTS) Signal
|
998 | Channel" registry <xref target="Key-Map"/>.</t>
|
999 | <table anchor="table2">
|
1000 | <thead>
|
1001 | <tr>
|
1002 | <th>Parameter Name</th>
|
1003 | <th>CBOR Key Value</th>
|
1004 | <th>CBOR Major Type</th>
|
1005 | <th>Change Controller</th>
|
1006 | <th>Specification Document(s)</th>
|
1007 | </tr>
|
1008 | </thead>
|
1009 | <tbody>
|
1010 | <tr>
|
1011 | <td>activation-type</td>
|
1012 | <td>52</td>
|
1013 | <td>0</td>
|
1014 | <td>IESG</td>
|
1015 | <td>RFC 9133</td>
|
1016 | </tr>
|
1017 | <tr>
|
1018 | <td>ietf-dots-signal-control:acl-list</td>
|
1019 | <td>53</td>
|
1020 | <td>4</td>
|
1021 | <td>IESG</td>
|
1022 | <td>RFC 9133</td>
|
1023 | </tr>
|
1024 | </tbody>
|
1025 | </table>
|
1026 | </section>
|
1027 | <section anchor="yang-iana">
|
1028 | <name>A New YANG Module</name>
|
1029 | <t>IANA has registered the following URI in the
|
1030 | "ns" subregistry within the "IETF XML Registry" <xref target="RFC3688"/>:</t>
|
1031 | <dl spacing="compact">
|
1032 | <dt>URI:</dt>
|
1033 | <dd>urn:ietf:params:xml:ns:yang:ietf-dots-signal-control</dd>
|
1034 | <dt>Registrant Contact:</dt>
|
1035 | <dd>The IESG.</dd>
|
1036 | <dt>XML:</dt>
|
1037 | <dd>N/A; the requested URI is an XML namespace.</dd>
|
1038 | </dl>
|
1039 | <t>IANA has registered the following YANG module
|
1040 | in the "YANG Module Names" subregistry <xref target="RFC6020"/>
|
1041 | within the "YANG Parameters" registry.</t>
|
1042 | <dl spacing="compact">
|
1043 | <dt>Name:</dt>
|
1044 | <dd>ietf-dots-signal-control</dd>
|
1045 | <dt>Namespace:</dt>
|
1046 | <dd>urn:ietf:params:xml:ns:yang:ietf-dots-signal-control</dd>
|
1047 | <dt>Maintained by IANA:</dt>
|
1048 | <dd>N</dd>
|
1049 | <dt>Prefix:</dt>
|
1050 | <dd>dots-control</dd>
|
1051 | <dt>Reference:</dt>
|
1052 | <dd>RFC 9133</dd>
|
1053 | </dl>
|
1054 | </section>
|
1055 | </section>
|
1056 | <section anchor="security">
|
1057 | <name>Security Considerations</name>
|
1058 | <t>The security considerations for the DOTS signal channel protocol are
|
1059 | discussed in <xref target="RFC9132" section="11"/>,
|
1060 | while those for the DOTS data channel protocol are discussed in <xref target="RFC8783" section="10"/>. The following
|
1061 | discusses the security considerations that are specific to the DOTS
|
1062 | signal channel extension defined in this document.</t>
|
1063 | <t>This specification does not allow the creation of new filtering rules,
|
1064 | which is the responsibility of the DOTS data channel. DOTS client
|
1065 | domains should be adequately prepared prior to an attack, e.g., by
|
1066 | creating filters that will be activated on demand when an attack is
|
1067 | detected.</t>
|
1068 | <t>A DOTS client is entitled to access only the resources it creates. In
|
1069 | particular, a DOTS client can not tweak filtering rules created by other
|
1070 | DOTS clients of the same DOTS client domain. As a reminder, DOTS servers
|
1071 | must associate filtering rules with the DOTS client that created these
|
1072 | resources. Failure to ensure such association by a DOTS server will have
|
1073 | severe impact on DOTS client domains.</t>
|
1074 | <t>A compromised DOTS client can use the filtering control capability to
|
1075 | exacerbate an ongoing attack. Likewise, such a compromised DOTS client
|
1076 | may abstain from reacting to an ACL conflict notification received from
|
1077 | the DOTS server during attacks. These are not new attack vectors, but
|
1078 | variations of threats discussed in <xref target="RFC9132"/> and <xref target="RFC8783"/>. DOTS
|
1079 | operators should carefully monitor and audit DOTS agents to detect
|
1080 | misbehaviors and deter misuses.</t>
|
1081 | </section>
|
1082 | </middle>
|
1083 | <back>
|
1084 | <references>
|
1085 | <name>References</name>
|
1086 | <references>
|
1087 | <name>Normative References</name>
|
1088 | <reference anchor="RFC2119" target="https://www.rfc-editor.org/info/rfc2119" derivedAnchor="RFC2119">
|
1089 | <front>
|
1090 | <title>Key words for use in RFCs to Indicate Requirement Levels</title>
|
1091 | <author initials="S." surname="Bradner" fullname="S. Bradner">
|
1092 | <organization/>
|
1093 | </author>
|
1094 | <date year="1997" month="March"/>
|
1095 | <abstract>
|
1096 | <t>In many standards track documents several words are used to signify the requirements in the specification. These words are often capitalized. This document defines these words as they should be interpreted in IETF documents. This document specifies an Internet Best Current Practices for the Internet Community, and requests discussion and suggestions for improvements.</t>
|
1097 | </abstract>
|
1098 | </front>
|
1099 | <seriesInfo name="BCP" value="14"/>
|
1100 | <seriesInfo name="RFC" value="2119"/>
|
1101 | <seriesInfo name="DOI" value="10.17487/RFC2119"/>
|
1102 | </reference>
|
1103 | <reference anchor="RFC3688" target="https://www.rfc-editor.org/info/rfc3688" derivedAnchor="RFC3688">
|
1104 | <front>
|
1105 | <title>The IETF XML Registry</title>
|
1106 | <author initials="M." surname="Mealling" fullname="M. Mealling">
|
1107 | <organization/>
|
1108 | </author>
|
1109 | <date year="2004" month="January"/>
|
1110 | <abstract>
|
1111 | <t>This document describes an IANA maintained registry for IETF standards which use Extensible Markup Language (XML) related items such as Namespaces, Document Type Declarations (DTDs), Schemas, and Resource Description Framework (RDF) Schemas.</t>
|
1112 | </abstract>
|
1113 | </front>
|
1114 | <seriesInfo name="BCP" value="81"/>
|
1115 | <seriesInfo name="RFC" value="3688"/>
|
1116 | <seriesInfo name="DOI" value="10.17487/RFC3688"/>
|
1117 | </reference>
|
1118 | <reference anchor="RFC6020" target="https://www.rfc-editor.org/info/rfc6020" derivedAnchor="RFC6020">
|
1119 | <front>
|
1120 | <title>YANG - A Data Modeling Language for the Network Configuration Protocol (NETCONF)</title>
|
1121 | <author initials="M." surname="Bjorklund" fullname="M. Bjorklund" role="editor">
|
1122 | <organization/>
|
1123 | </author>
|
1124 | <date year="2010" month="October"/>
|
1125 | <abstract>
|
1126 | <t>YANG is a data modeling language used to model configuration and state data manipulated by the Network Configuration Protocol (NETCONF), NETCONF remote procedure calls, and NETCONF notifications. [STANDARDS-TRACK]</t>
|
1127 | </abstract>
|
1128 | </front>
|
1129 | <seriesInfo name="RFC" value="6020"/>
|
1130 | <seriesInfo name="DOI" value="10.17487/RFC6020"/>
|
1131 | </reference>
|
1132 | <reference anchor="RFC7950" target="https://www.rfc-editor.org/info/rfc7950" derivedAnchor="RFC7950">
|
1133 | <front>
|
1134 | <title>The YANG 1.1 Data Modeling Language</title>
|
1135 | <author initials="M." surname="Bjorklund" fullname="M. Bjorklund" role="editor">
|
1136 | <organization/>
|
1137 | </author>
|
1138 | <date year="2016" month="August"/>
|
1139 | <abstract>
|
1140 | <t>YANG is a data modeling language used to model configuration data, state data, Remote Procedure Calls, and notifications for network management protocols. This document describes the syntax and semantics of version 1.1 of the YANG language. YANG version 1.1 is a maintenance release of the YANG language, addressing ambiguities and defects in the original specification. There are a small number of backward incompatibilities from YANG version 1. This document also specifies the YANG mappings to the Network Configuration Protocol (NETCONF).</t>
|
1141 | </abstract>
|
1142 | </front>
|
1143 | <seriesInfo name="RFC" value="7950"/>
|
1144 | <seriesInfo name="DOI" value="10.17487/RFC7950"/>
|
1145 | </reference>
|
1146 | <reference anchor="RFC8174" target="https://www.rfc-editor.org/info/rfc8174" derivedAnchor="RFC8174">
|
1147 | <front>
|
1148 | <title>Ambiguity of Uppercase vs Lowercase in RFC 2119 Key Words</title>
|
1149 | <author initials="B." surname="Leiba" fullname="B. Leiba">
|
1150 | <organization/>
|
1151 | </author>
|
1152 | <date year="2017" month="May"/>
|
1153 | <abstract>
|
1154 | <t>RFC 2119 specifies common key words that may be used in protocol specifications. This document aims to reduce the ambiguity by clarifying that only UPPERCASE usage of the key words have the defined special meanings.</t>
|
1155 | </abstract>
|
1156 | </front>
|
1157 | <seriesInfo name="BCP" value="14"/>
|
1158 | <seriesInfo name="RFC" value="8174"/>
|
1159 | <seriesInfo name="DOI" value="10.17487/RFC8174"/>
|
1160 | </reference>
|
1161 | <reference anchor="RFC8783" target="https://www.rfc-editor.org/info/rfc8783" derivedAnchor="RFC8783">
|
1162 | <front>
|
1163 | <title>Distributed Denial-of-Service Open Threat Signaling (DOTS) Data Channel Specification</title>
|
1164 | <author initials="M." surname="Boucadair" fullname="M. Boucadair" role="editor">
|
1165 | <organization/>
|
1166 | </author>
|
1167 | <author initials="T." surname="Reddy.K" fullname="T. Reddy.K" role="editor">
|
1168 | <organization/>
|
1169 | </author>
|
1170 | <date year="2020" month="May"/>
|
1171 | <abstract>
|
1172 | <t>The document specifies a Distributed Denial-of-Service Open Threat Signaling (DOTS) data channel used for bulk exchange of data that cannot easily or appropriately communicated through the DOTS signal channel under attack conditions.</t>
|
1173 | <t>This is a companion document to "Distributed Denial-of-Service Open Threat Signaling (DOTS) Signal Channel Specification" (RFC 8782).</t>
|
1174 | </abstract>
|
1175 | </front>
|
1176 | <seriesInfo name="RFC" value="8783"/>
|
1177 | <seriesInfo name="DOI" value="10.17487/RFC8783"/>
|
1178 | </reference>
|
1179 | <reference anchor="RFC8791" target="https://www.rfc-editor.org/info/rfc8791" derivedAnchor="RFC8791">
|
1180 | <front>
|
1181 | <title>YANG Data Structure Extensions</title>
|
1182 | <author initials="A." surname="Bierman" fullname="A. Bierman">
|
1183 | <organization/>
|
1184 | </author>
|
1185 | <author initials="M." surname="Björklund" fullname="M. Björklund">
|
1186 | <organization/>
|
1187 | </author>
|
1188 | <author initials="K." surname="Watsen" fullname="K. Watsen">
|
1189 | <organization/>
|
1190 | </author>
|
1191 | <date year="2020" month="June"/>
|
1192 | <abstract>
|
1193 | <t>This document describes YANG mechanisms for defining abstract data structures with YANG.</t>
|
1194 | </abstract>
|
1195 | </front>
|
1196 | <seriesInfo name="RFC" value="8791"/>
|
1197 | <seriesInfo name="DOI" value="10.17487/RFC8791"/>
|
1198 | </reference>
|
1199 | <reference anchor="RFC9132" target="https://www.rfc-editor.org/info/rfc9132" derivedAnchor="RFC9132">
|
1200 | <front>
|
1201 | <title>Distributed Denial-of-Service Open Threat Signaling (DOTS) Signal Channel Specification</title>
|
1202 | <author initials="M" surname="Boucadair" fullname="Mohamed Boucadair" role="editor">
|
1203 | <organization/>
|
1204 | </author>
|
1205 | <author initials="J" surname="Shallow" fullname="Jon Shallow">
|
1206 | <organization/>
|
1207 | </author>
|
1208 | <author initials="T" surname="Reddy.K" fullname="Tirumaleswar Reddy.K">
|
1209 | <organization/>
|
1210 | </author>
|
1211 | <date month="September" year="2021"/>
|
1212 | </front>
|
1213 | <seriesInfo name="RFC" value="9132"/>
|
1214 | <seriesInfo name="DOI" value="10.17487/RFC9132"/>
|
1215 | </reference>
|
1216 | </references>
|
1217 | <references>
|
1218 | <name>Informative References</name>
|
1219 | <reference anchor="INTEROP" target="https://datatracker.ietf.org/meeting/103/materials/slides-103-dots-interop-report-from-ietf-103-hackathon-00" derivedAnchor="INTEROP">
|
1220 | <front>
|
1221 | <title>DOTS Interop test report, IETF 103 Hackathon</title>
|
1222 | <author fullname="Kaname Nishizuka" initials="K." surname="Nishizuka">
|
1223 | <organization>NTT Communications</organization>
|
1224 | <address>
|
1225 | <postal>
|
1226 | <street>GranPark 16F 3-4-1 Shibaura, Minato-ku</street>
|
1227 | <city>Tokyo</city>
|
1228 | <region/>
|
1229 | <code>108-8118</code>
|
1230 | <country>Japan</country>
|
1231 | </postal>
|
1232 | <email>kaname@nttv6.jp</email>
|
1233 | </address>
|
1234 | </author>
|
1235 | <author fullname="Jon Shallow" initials="J." surname="Shallow">
|
1236 | <organization>J.NCC Group</organization>
|
1237 | <address>
|
1238 | <postal>
|
1239 | <street/>
|
1240 | <city/>
|
1241 | <region/>
|
1242 | <code/>
|
1243 | <country/>
|
1244 | </postal>
|
1245 | <phone/>
|
1246 | <email/>
|
1247 | <uri/>
|
1248 | </address>
|
1249 | </author>
|
1250 | <author fullname="Liang Xia" initials="L." surname="Xia">
|
1251 | <organization>Huawei</organization>
|
1252 | <address>
|
1253 | <postal>
|
1254 | <street/>
|
1255 | <city/>
|
1256 | <region/>
|
1257 | <code/>
|
1258 | <country/>
|
1259 | </postal>
|
1260 | <phone/>
|
1261 | <email/>
|
1262 | <uri/>
|
1263 | </address>
|
1264 | </author>
|
1265 | <date month="November" year="2018"/>
|
1266 | </front>
|
1267 | </reference>
|
1268 | <reference anchor="Key-Map" target="https://www.iana.org/assignments/dots" derivedAnchor="Key-Map">
|
1269 | <front>
|
1270 | <title>Distributed Denial-of-Service Open Threat Signaling (DOTS) Signal Channel</title>
|
1271 | <author fullname="IANA">
|
1272 | <organization/>
|
1273 | </author>
|
1274 | </front>
|
1275 | </reference>
|
1276 | <reference anchor="RFC7951" target="https://www.rfc-editor.org/info/rfc7951" derivedAnchor="RFC7951">
|
1277 | <front>
|
1278 | <title>JSON Encoding of Data Modeled with YANG</title>
|
1279 | <author initials="L." surname="Lhotka" fullname="L. Lhotka">
|
1280 | <organization/>
|
1281 | </author>
|
1282 | <date year="2016" month="August"/>
|
1283 | <abstract>
|
1284 | <t>This document defines encoding rules for representing configuration data, state data, parameters of Remote Procedure Call (RPC) operations or actions, and notifications defined using YANG as JavaScript Object Notation (JSON) text.</t>
|
1285 | </abstract>
|
1286 | </front>
|
1287 | <seriesInfo name="RFC" value="7951"/>
|
1288 | <seriesInfo name="DOI" value="10.17487/RFC7951"/>
|
1289 | </reference>
|
1290 | <reference anchor="RFC8340" target="https://www.rfc-editor.org/info/rfc8340" derivedAnchor="RFC8340">
|
1291 | <front>
|
1292 | <title>YANG Tree Diagrams</title>
|
1293 | <author initials="M." surname="Bjorklund" fullname="M. Bjorklund">
|
1294 | <organization/>
|
1295 | </author>
|
1296 | <author initials="L." surname="Berger" fullname="L. Berger" role="editor">
|
1297 | <organization/>
|
1298 | </author>
|
1299 | <date year="2018" month="March"/>
|
1300 | <abstract>
|
1301 | <t>This document captures the current syntax used in YANG module tree diagrams. The purpose of this document is to provide a single location for this definition. This syntax may be updated from time to time based on the evolution of the YANG language.</t>
|
1302 | </abstract>
|
1303 | </front>
|
1304 | <seriesInfo name="BCP" value="215"/>
|
1305 | <seriesInfo name="RFC" value="8340"/>
|
1306 | <seriesInfo name="DOI" value="10.17487/RFC8340"/>
|
1307 | </reference>
|
1308 | <reference anchor="RFC8612" target="https://www.rfc-editor.org/info/rfc8612" derivedAnchor="RFC8612">
|
1309 | <front>
|
1310 | <title>DDoS Open Threat Signaling (DOTS) Requirements</title>
|
1311 | <author initials="A." surname="Mortensen" fullname="A. Mortensen">
|
1312 | <organization/>
|
1313 | </author>
|
1314 | <author initials="T." surname="Reddy" fullname="T. Reddy">
|
1315 | <organization/>
|
1316 | </author>
|
1317 | <author initials="R." surname="Moskowitz" fullname="R. Moskowitz">
|
1318 | <organization/>
|
1319 | </author>
|
1320 | <date year="2019" month="May"/>
|
1321 | <abstract>
|
1322 | <t>This document defines the requirements for the Distributed Denial-of- Service (DDoS) Open Threat Signaling (DOTS) protocols enabling coordinated response to DDoS attacks.</t>
|
1323 | </abstract>
|
1324 | </front>
|
1325 | <seriesInfo name="RFC" value="8612"/>
|
1326 | <seriesInfo name="DOI" value="10.17487/RFC8612"/>
|
1327 | </reference>
|
1328 | <reference anchor="RFC8811" target="https://www.rfc-editor.org/info/rfc8811" derivedAnchor="RFC8811">
|
1329 | <front>
|
1330 | <title>DDoS Open Threat Signaling (DOTS) Architecture</title>
|
1331 | <author initials="A." surname="Mortensen" fullname="A. Mortensen" role="editor">
|
1332 | <organization/>
|
1333 | </author>
|
1334 | <author initials="T." surname="Reddy.K" fullname="T. Reddy.K" role="editor">
|
1335 | <organization/>
|
1336 | </author>
|
1337 | <author initials="F." surname="Andreasen" fullname="F. Andreasen">
|
1338 | <organization/>
|
1339 | </author>
|
1340 | <author initials="N." surname="Teague" fullname="N. Teague">
|
1341 | <organization/>
|
1342 | </author>
|
1343 | <author initials="R." surname="Compton" fullname="R. Compton">
|
1344 | <organization/>
|
1345 | </author>
|
1346 | <date year="2020" month="August"/>
|
1347 | <abstract>
|
1348 | <t>This document describes an architecture for establishing and maintaining Distributed Denial-of-Service (DDoS) Open Threat Signaling (DOTS) within and between domains. The document does not specify protocols or protocol extensions, instead focusing on defining architectural relationships, components, and concepts used in a DOTS deployment.</t>
|
1349 | </abstract>
|
1350 | </front>
|
1351 | <seriesInfo name="RFC" value="8811"/>
|
1352 | <seriesInfo name="DOI" value="10.17487/RFC8811"/>
|
1353 | </reference>
|
1354 | </references>
|
1355 | </references>
|
1356 | <section anchor="ack" numbered="false">
|
1357 | <name>Acknowledgements</name>
|
1358 | <t>Many thanks to <contact fullname="Wei Pan"/>, <contact fullname="Xia Liang"/>, <contact fullname="Jon Shallow"/>, <contact fullname="Dan Wing"/>, <contact fullname="Christer Holmberg"/>, <contact fullname="Shawn Emery"/>, <contact fullname="Tim Chown"/>, <contact fullname="Murray Kucherawy"/>, <contact fullname="Roman Danyliw"/>, <contact fullname="Erik Kline"/>, and <contact fullname="Éric Vyncke"/> for the comments.</t>
|
1359 | <t>Thanks to <contact fullname="Benjamin Kaduk"/> for the AD review.</t>
|
1360 | </section>
|
1361 | </back>
|
1362 | </rfc>
|