idnits 2.17.1 draft-ietf-rtcweb-qos-00.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- == It seems as if not all pages are separated by form feeds - found 0 form feeds but 10 pages Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- == There are 2 instances of lines with non-RFC2606-compliant FQDNs in the document. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (October 15, 2012) is 4212 days in the past. Is this intentional? -- Found something which looks like a code comment -- if you have code sections in the document, please surround them with '' and '' lines. Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Unused Reference: '1' is defined on line 364, but no explicit reference was found in the text == Unused Reference: '2' is defined on line 367, but no explicit reference was found in the text == Unused Reference: '3' is defined on line 372, but no explicit reference was found in the text == Unused Reference: '4' is defined on line 376, but no explicit reference was found in the text == Unused Reference: '5' is defined on line 379, but no explicit reference was found in the text ** Downref: Normative reference to an Informational RFC: RFC 4594 (ref. '1') -- Possible downref: Non-RFC (?) normative reference: ref. '6' Summary: 1 error (**), 0 flaws (~~), 8 warnings (==), 3 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group S. Dhesikan 3 Internet-Draft Cisco 4 Intended status: Standards Track D. Druta, Ed. 5 Expires: April 15, 2013 ATT 6 P. Jones 7 J. Polk 8 Cisco 9 October 15, 2012 11 DSCP and other packet markings for RTCWeb QoS 12 draft-ietf-rtcweb-qos-00 14 Abstract 16 Many networks, such as Service Provider and Enterprise networks, can 17 provide per packet treatments based on Differentiated Services Code 18 Points (DSCP) on a per hop basis. This document defines the 19 recommended DSCP values for browsers to use for various classes of 20 traffic. 22 This draft is a very early and far from done. It is meant to provide 23 the structure for the idea of how to do this but much discussion is 24 needed about the details. 26 Status of this Memo 28 This Internet-Draft is submitted in full conformance with the 29 provisions of BCP 78 and BCP 79. 31 Internet-Drafts are working documents of the Internet Engineering 32 Task Force (IETF). Note that other groups may also distribute 33 working documents as Internet-Drafts. The list of current Internet- 34 Drafts is at http://datatracker.ietf.org/drafts/current/. 36 Internet-Drafts are draft documents valid for a maximum of six 37 months and may be updated, replaced, or obsoleted by other documents 38 at any time. It is inappropriate to use Internet-Drafts as 39 reference material or to cite them other than as "work in progress." 41 This Internet-Draft will expire on April 15, 2013. 43 Copyright Notice 45 Copyright (c) 2012 IETF Trust and the persons identified as the 46 document authors. All rights reserved. 48 This document is subject to BCP 78 and the IETF Trust's Legal 49 Provisions Relating to IETF Documents 50 (http://trustee.ietf.org/license-info) in effect on the date of 51 publication of this document. Please review these documents 52 carefully, as they describe your rights and restrictions with 53 respect to this document. Code Components extracted from this 54 document must include Simplified BSD License text as described in 55 Section 4.e of the Trust Legal Provisions and are provided without 56 warranty as described in the Simplified BSD License. 58 Table of Contents 60 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 61 2. Relation to Other Standards . . . . . . . . . . . . . . . . . . 3 62 3. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . 4 63 4. Inputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 64 5. DSCP Mappings . . . . . . . . . . . . . . . . . . . . . . . . . 4 65 6. QCI Mapping . . . . . . . . . . . . . . . . . . . . . . . . . . 5 66 7. WiFI Mapping . . . . . . . . . . . . . . . . . . . . . . . . . 5 67 8. W3C API Implications . . . . . . . . . . . . . . . . . . . . . 6 68 9. Security Considerations . . . . . . . . . . . . . . . . . . . . 6 69 10. IANA Considerations . . . . . . . . . . . . . . . . . . . . . . 6 70 11. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 6 71 12. Appendix: Code Hints . . . . . . . . . . . . . . . . . . . . . 7 72 13. Normative References . . . . . . . . . . . . . . . . . . . . . 9 73 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 9 75 1. Introduction 77 DiffServ style packet marking can help provide QoS in some 78 environments. There are many use cases where such marking does not 79 help, but it seldom makes things worse, if packets are marked 80 appropriately. In other words, when attempting to avoid congestion 81 by marking certain traffic flows, say all audio or all audio and 82 video, marking too many audio and/or video flows for a given 83 network's capacity can prevent desirable results. Either too much 84 other traffic will be starved, or there is not enough capacity for 85 the preferentially marked packets (i.e., audio and/or video). 87 This draft proposes how a browser and other VoIP applications can 88 mark packets. This draft does not contradict or redefine any advice 89 from previous IETF RFCs but simply provides a simple set of 90 recommendations for implementors based on the previous RFCs. 92 There are some environments where priority markings frequently help. 93 These include: 95 1. If the congested link is the broadband uplink in a Cable or DSL 96 scenario, often residential routers/NAT support preferential 97 treatment based on DSCP. 99 2. If the congested link is a local WiFi network, marking may help. 101 3. In some cellular style deployments, markings may help in cases 102 where the network does not remove them. 104 Traditionally DSCP values have been thought of as being site 105 specific, with each site selecting its own code points for each QoS 106 level. However in the RTCWeb use cases, the browsers need to set 107 them to something when there is no site specific information. This 108 document describes a reasonable default set of DSCP code point values 109 drawn from existing RFCs and common usage. These code points are 110 solely defaults. Future drafts may define mechanisms for site 111 specific mappings to override the values provided in this draft. 113 This draft defines some inputs that the browser can look at to 114 determine how to set the various packet markings and defines the a 115 mapping from abstract QoS policies (media type, priority level) to 116 those packet markings. 118 2. Relation to Other Standards 120 This specification does not change or override the advice in any 121 other standards about setting packet markings. It simply provides 122 a non-normative summary of them and provides the context of how they 123 relate into the RTCWeb context. It also specified the requirements 124 for the W3C API to understand what it needs to control and how the 125 control splits between things the JavaScript application running in 126 the browser can control and things the browser needs to control. In 127 some cases, such as DSCP where the normative RFC leaves open multiple 128 options to choose from, this clarifies which choice should be used in 129 the RTCWeb context. 131 3. Terminology 133 TODO - add the boiler plate 135 4. Inputs 137 The first input is the type of the media. The browser provides this 138 input as it knows if the media is audio, video, or data. In this 139 specification, both interactive and streaming media is included. 140 They are treated in different categories as their QoS requirements 141 are slightly different. The second input is the relative treatment 142 of the stream within that session. Many applications have multiple 143 video streams and often some are more important than others. 144 JavaScript applications can tell the browser whether a particular 145 media stream is high, medium, or low importance to the application. 147 5. DSCP Mappings 149 Below is a table of DSCP markings for each application type RTCWeb 150 is interested in. These DSCPs for each application type listed are a 151 reasonable default set of code point values, and currently not 152 mandatory for every usage. For example, some networks may have a 153 policy in place to have Interactive Video use the EF DSCP. 155 +-----------------------+-----------+-----------+-----------+ 156 | Application Type | Low | Medium | High | 157 +-----------------------+-----------+-----------+-----------+ 158 | Audio | 46 (EF) | 46 (EF) | 46 (EF) | 159 | Interactive Video | 38 (AF43) | 36 (AF42) | 34 (AF41) | 160 | Non-Interactive Video | 26 (AF33) | 28 (AF32) | 30 (AF31) | 161 | Data | 8 (CS1) | 0 (BE) | 10 (AF11) | 162 +-----------------------+-----------+-----------+-----------+ 164 Table 1 166 [Editor's Note: the application type is currently inconsistent with 167 similar applications defined in [6]. Further 168 discussion is likely needed to resolve this.] 170 6. QCI Mapping 172 +-----------------------+-----+--------+------+ 173 | Application Type | Low | Medium | High | 174 +-----------------------+-----+--------+------+ 175 | Audio | 1 | 1 | 1 | 176 | Interactive Video | 2 | 2 | 2 | 177 | Non-Interactive Video | 8 | 6 | 4 | 178 | Data | 9 | 9 | 3 | 179 +-----------------------+-----+--------+------+ 181 Table 2 183 This corresponds to the mapping provided in TODO REF which are: QCI 184 values (LTE) 186 +-------+--------+-----+--------------------------------------------+ 187 | Value | | | Use | 188 +-------+--------+-----+--------------------------------------------+ 189 | 1 | GBR | 2 | Interactive Voice | 190 | 2 | GBR | 4 | Interactive Video | 191 | 3 | GBR | 5 | Non-Interactive Video | 192 | 4 | GBR | 3 | Real Time Gaming | 193 | 5 | Non-BG | R 1 | IMS Signalling | 194 | 6 | Non-BG | R 7 | interactive Voice, video, games | 195 | 7-9 | Non-BG | R 6 | non interactive video / TCP web, email, / | 196 | | | | Platinum vs gold user | 197 +-------+--------+-----+--------------------------------------------+ 199 Table 3 201 7. WiFI Mapping 203 +-----------------------+-----+--------+------+ 204 | | Low | Medium | High | 205 +-----------------------+-----+--------+------+ 206 | Audio | 6 | 6 | 6 | 207 | Interactive Video | 5 | 5 | 5 | 208 | Non-Interactive Video | 4 | 4 | 4 | 209 | Data | 1 | 0 | 3 | 210 +-----------------------+-----+--------+------+ 212 Table 4 214 This corresponds to the mappings from TODO REF of 215 +-------+----+------------------+---------------------+-------------+ 216 | Value | | Traffic Type | Access Category | Designation | 217 | | | | (AC) | | 218 +-------+----+------------------+---------------------+-------------+ 219 | 1 | BK | Background | AC_BK | Background | 220 | 2 | - | (spare) | AC_BK | Background | 221 | 0 | BE | Best Effort | AC_BE | Best Effort | 222 | 3 | EE | Excellent Effort | AC_BE | Best Effort | 223 | 4 | CL | Controlled Load | AC_VI | Video | 224 | 5 | VI | Video | AC_VI | Video | 225 | 6 | VO | Voice | AC_VO | Voice | 226 | 7 | NC | Network Control | AC_VO | Voice | 227 +-------+----+------------------+---------------------+-------------+ 229 Table 5 231 8. W3C API Implications 233 To work with this proposal, the W3C specification would need to 234 provide a way to specify the importance of media and data streams. 236 The W3C API should also provide a way for the application to find out 237 the source and destination IP and ports of any flow as well as the 238 DSCP value or other markings in use for that flow. The JavaScript 239 application can then communicate this to a web service that may 240 install a particular policy for that flow. 242 [Editor's Note: the idea of bundling applications/media needs to be 243 further explored.] 245 9. Security Considerations 247 TODO - discuss implications of what browser can set and what 248 JavaScript can set 250 10. IANA Considerations 252 This specification does not require any actions from IANA. 254 11. Acknowledgements 256 Cullen Jennings was one of the authors of this text in the original 257 individual submission but was unceremoniously kicked off by the 258 chairs when it became a WG version. Thanks for hints on code to do 259 this from Paolo Severini, Jim Hasselbrook, Joe Marcus, and Erik 260 Nordmark. 262 12. Appendix: Code Hints 264 On windows setting the source interface works but BSD, OSX, Linux use 265 weak end-system model and will route out different interface if that 266 looks like a better route. (TODO - Can someone verify this with 267 specific versions?) 269 In windows you might be able to tell something about priority of an 270 interface for ICE purposes with WlanQueryInterface or GetIfTable. 272 The specific mechanisms required to set DSCP code points depend on 273 the application platform. 275 In windows, setting the DSCP is not easy. See Knowledge Base Article 276 KB248611. TODO - add more information about what can be done for 277 windows. 279 For most unix variants, the following program can set DSCP. 281 TODO - make this work in V6. For v6 have a look at IPv6_TCLASS or 282 better the tclass part of sin6_flowid for IPv6 284 TODO - Can someone test and report back results of program in iOS, 285 Android, Linux, OSX, BSD. 287 Example test program: 289 #include 290 #include 291 #include 292 #include 293 #include 294 #include 295 #include 296 #include 297 #include 298 #include 300 #define MSG "Hello, World!" 302 int 303 main(void) { 304 int sock = -1; 305 struct sockaddr *local_addr = NULL; 306 struct sockaddr_in sockin, host; 307 int tos = 0x60; /* CS3 */ 308 socklen_t socksiz = 0; 309 char *buffer = NULL; 311 sock = socket(AF_INET, SOCK_DGRAM, 0); 312 if (sock < 0) { 313 fprintf(stderr,"Error: %s\n", strerror(errno)); 314 exit(-1); 315 } 317 memset(&sockin, 0, sizeof(sockin)); 318 sockin.sin_family = PF_INET; 319 sockin.sin_addr.s_addr = inet_addr("11.1.1.1"); 320 socksiz = sizeof(sockin); 322 local_addr = (struct sockaddr *) &sockin; 324 /* Set ToS/DSCP */ 325 if (setsockopt(sock, IPPROTO_IP, IP_TOS, &tos, 326 sizeof(tos)) < 0) { 327 fprintf(stderr,"Error setting TOS: %s\n", strerror(errno)); 328 } 330 /* Bind to a specific local address */ 331 if (bind(sock, local_addr, socksiz) < 0) { 332 fprintf(stderr,"Error binding to socket: %s\n", strerror(errno)); 333 close(sock); sock=-1; 334 exit(-1); 335 } 337 buffer = (char *) malloc(strlen(MSG) + 1); 338 if ( buffer == NULL ) { 339 fprintf(stderr,"Error allocating memory: %s\n", strerror(errno)); 340 close( sock ); sock=-1; 341 exit(-1); 342 } 343 strlcpy(buffer, MSG, strlen(MSG) + 1); 344 memset(&host, 0, sizeof(host)); 345 host.sin_family = PF_INET; 346 host.sin_addr.s_addr = inet_addr("10.1.1.1"); 347 host.sin_port = htons(12345); 349 if (sendto(sock, buffer, strlen(buffer), 0, 350 (struct sockaddr *) &host, sizeof(host)) < 0) { 351 fprintf(stderr,"Error sending message: %s\n", strerror(errno)); 352 close(sock); sock=-1; 353 free(buffer); buffer=NULL; 354 exit(-1); 355 } 356 free(buffer); buffer=NULL; 357 close(sock); sock=-1; 359 return 0; 360 } 362 13. Normative References 364 [1] Babiarz, J., Chan, K., and F. Baker, "Configuration Guidelines 365 for DiffServ Service Classes", RFC 4594, August 2006. 367 [2] Davie, B., Charny, A., Bennet, J., Benson, K., Le Boudec, J., 368 Courtney, W., Davari, S., Firoiu, V., and D. Stiliadis, "An 369 Expedited Forwarding PHB (Per-Hop Behavior)", RFC 3246, 370 March 2002. 372 [3] Nichols, K., Blake, S., Baker, F., and D. Black, "Definition of 373 the Differentiated Services Field (DS Field) in the IPv4 and 374 IPv6 Headers", RFC 2474, December 1998. 376 [4] Heinanen, J., Baker, F., Weiss, W., and J. Wroclawski, "Assured 377 Forwarding PHB Group", RFC 2597, June 1999. 379 [5] Bradner, S., "Key words for use in RFCs to Indicate Requirement 380 Levels", BCP 14, RFC 2119, March 1997. 382 [6] Polk, J., Dhesikan, S., "The Session Description Protocol (SDP) 383 'trafficclass' Attribute", "work in progress", July 2012 385 Authors' Addresses 387 Subha Dhesikan 388 Cisco 390 Email: sdhesika@cisco.com 392 Dan Druta (editor) 393 ATT 395 Email: dd5826@att.com 397 Paul Jones 398 Cisco 400 Email: paulej@packetizer.com 401 James Polk 402 Cisco 404 Email: jmpolk@cisco.com