idnits 2.17.00 (12 Aug 2021) /tmp/idnits53837/draft-wkumari-dhc-capport-10.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (January 28, 2015) is 2670 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Missing Reference: 'RFC2939' is mentioned on line 351, but not defined Summary: 0 errors (**), 0 flaws (~~), 2 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group W. Kumari 3 Internet-Draft Google 4 Intended status: Informational O. Gudmundsson 5 Expires: August 1, 2015 Shinkuro Inc. 6 P. Ebersman 7 Comcast 8 S. Sheng 9 ICANN 10 January 28, 2015 12 Captive-Portal identification in DHCP / RA 13 draft-wkumari-dhc-capport-10 15 Abstract 17 In many environments offering short-term or temporary Internet access 18 (such as coffee shops), it is common to start new connections in a 19 captive portal mode. This highly restricts what the customer can do 20 until the customer has authenticated. 22 This document describes a DHCP option (and a RA extension) to inform 23 clients that they are behind some sort of captive portal device, and 24 that they will need to authenticate to get Internet Access. 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 months 37 and may be updated, replaced, or obsoleted by other documents at any 38 time. It is inappropriate to use Internet-Drafts as reference 39 material or to cite them other than as "work in progress." 41 This Internet-Draft will expire on August 1, 2015. 43 Copyright Notice 45 Copyright (c) 2015 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 respect 53 to this document. Code Components extracted from this document must 54 include Simplified BSD License text as described in Section 4.e of 55 the Trust Legal Provisions and are provided without warranty as 56 described in the Simplified BSD License. 58 Table of Contents 60 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 61 1.1. Requirements notation . . . . . . . . . . . . . . . . . . 3 62 2. Background . . . . . . . . . . . . . . . . . . . . . . . . . 3 63 2.1. DNS Redirection . . . . . . . . . . . . . . . . . . . . . 4 64 2.2. HTTP Redirection . . . . . . . . . . . . . . . . . . . . 4 65 2.3. IP Hijacking . . . . . . . . . . . . . . . . . . . . . . 4 66 3. The Captive-Portal Option . . . . . . . . . . . . . . . . . . 5 67 3.1. IPv4 DHCP Option . . . . . . . . . . . . . . . . . . . . 5 68 3.2. IPv6 DHCP Option . . . . . . . . . . . . . . . . . . . . 6 69 4. The Captive-Portal IPv6 RA Option . . . . . . . . . . . . . . 6 70 5. Use of the Captive-Portal Option . . . . . . . . . . . . . . 7 71 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 8 72 7. Security Considerations . . . . . . . . . . . . . . . . . . . 8 73 8. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 9 74 9. Normative References . . . . . . . . . . . . . . . . . . . . 9 75 Appendix A. Changes / Author Notes. . . . . . . . . . . . . . . 9 76 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 11 78 1. Introduction 80 In many environments, users need to connect to a captive portal 81 device and agree to an acceptable use policy and / or provide billing 82 information before they can access the Internet. 84 Many devices perform DNS, HTTP, and / or IP hijacks in order to 85 present the user with the captive portal web page. These workarounds 86 and techniques resemble attacks that DNSSEC and TLS are intended to 87 protect against. This document describe a DHCP option (Captive 88 Portal) and an IPv6 Router Advertisement (RA) extension that informs 89 clients that they are behind a captive portal device and how to 90 contact it. 92 This document neither condones nor condemns the use of captive 93 portals; instead, it recognises that their apparent necessity, and 94 attempts to improve the user experience. 96 [ Ed note: This solution is somewhat similar / complements 802.11u / 97 WiFi Passpoint Online Sign-up, but is much simpler, easier to deploy, 98 and works on wired as well ] 100 1.1. Requirements notation 102 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 103 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 104 document are to be interpreted as described in [RFC2119]. 106 2. Background 108 Some ISPs implement a captive portal (CP) - a system that intercepts 109 user requests and redirects them to an interstitial login page - in 110 order to require the user accept an Acceptable Use Policy (AUP), 111 provide billing information, or otherwise authenticate a user prior 112 to allowing them to access the Internet. 114 Captive portals intercept and redirect user requests in a number of 115 ways, including: 117 o DNS Redirection 119 o IP Redirection 121 o HTTP Redirection 123 o Restricted scope addresses 125 o Traffic blocking (until the user is authenticated) 127 In order to ensure that the user is unable to access the Internet 128 until they have satisfied the requirements, captive portals usually 129 implement IP based filters, or place the user into a restricted VLAN 130 (or restricted IP range) until after they have been authorized / 131 satisfied. 133 These techniques are very similar to attacks that protocols (such as 134 VPNs, DNSSEC, TLS) are designed to protect against. The interaction 135 of these protections and the interception leads to poor user 136 experiences, such as long timeouts, inability to reach the captive 137 portal web page, etc. The interception may also leak user 138 information (for example, if the captive portal intercepts and logs 139 an HTTP Cookie, or URL of the form http://fred:password@example.com). 140 The user is often unaware of what is causing the issue (their browser 141 appears to hang, saying something like "Downloading Proxy Script", or 142 simply "The Internet doesn't work"), and they become frustrated. 143 This may result in them not purchasing the Internet access provided 144 by the captive portal. The connectivity attempts may also facilitate 145 OS fingerprinting even before a client attempts to connect to the 146 portal itself. 148 2.1. DNS Redirection 150 The CP either intercepts all DNS traffic or advertises itself (for 151 example using DHCP) as the recursive server for the network. Until 152 the user has authenticated to the captive portal, the CP responds to 153 all DNS requests listing the address of its web portal. Once the 154 user has authenticated, the CP returns the "correct" addresses. 156 This technique has many shortcomings. It fails if the client is 157 performing DNSSEC validation, is running their own resolver, is using 158 a VPN, or already has the DNS information cached. 160 2.2. HTTP Redirection 162 In this implementation, the CP acts like a transparent HTTP proxy; 163 but when it sees an HTTP request from an unauthenticated client using 164 HTTP/1.0, it intercepts the request and responds with an HTTP status 165 code 302 to redirect the client to the Captive Portal Login. If the 166 client is using HTTP/1.1, we respond with a status code 303 See 167 Other. 169 This technique has a number of issues, including: 171 o It fails if the user is only using HTTPS. 173 o It exposes various private user information, such as HTTP Cookies, 174 etc. 176 o It doesn't work if the client has a VPN and / or proxies their web 177 traffic to an external web proxy. 179 2.3. IP Hijacking 181 In this scenario, the captive portal intercepts connections to any IP 182 address. It spoofs the destination IP address and "pretends" to be 183 whatever the user tried to access. 185 This technique has issues similar to the HTTP solution, but may also 186 break other protocols, and may expose more of the user's private 187 information. 189 3. The Captive-Portal Option 191 The Captive Portal DHCP / RA Option informs the client that it is 192 behind a captive portal and provides the URI to access an 193 authentication page. This is primarily intended to improve the user 194 experience; for the foreseeable future (until such time that most 195 systems implement this technique) captive portals will still need to 196 implement the interception techniques to serve legacy clients. 198 In order to support multiple "classes" of clients (e.g: IPv4 only, 199 IPv6 only with DHCPv6, IPv6 only with RA) the captive portal can 200 provide the URI via multiple methods (IPv4 DHCP, IPv6 DHCP, IPv6 RA). 201 The captive portal operator should ensure that the URIs handed out 202 are equivalent to reduce the chance of operational problems. 204 In order to avoid having to perform DNS interception, the URI SHOULD 205 contain an address literal, but MAY contain a DNS name if the captive 206 portal allows the client to perform DNS requests to resolve the name. 208 [ED NOTE: Using an address literal is less than ideal, but better 209 than the alternatives. Recommending a DNS name means that the CP 210 would need to allow DNS from unauthenticated clients (as we don't 211 want to force users to use the CP's provided DNS) and some users 212 would use this to DNS Tunnel out, which may make the CP admin block 213 external recursives). DNS is needed to allow operators to serve SSL/ 214 TLS for e.g billing (certificates with IP addresses are frowned upon 215 :-))] 217 3.1. IPv4 DHCP Option 219 The format of the IPv4 Captive-Portal DHCP option is shown below. 221 Code Len Data 222 +------+------+------+------+------+-- --+-----+ 223 | code | len | URI ... | 224 +------+------+------+------+------+-- --+-----+ 226 o Code: The Captive-Portal DHCPv4 Option (TBA1) 228 o Len: The length, in octets of the URI. 230 o URI: The URI of the authentication page that the user should 231 connect to. 233 3.2. IPv6 DHCP Option 235 The format of the IPv6 Captive-Portal DHCP option is shown below. 236 Other than the code it is identical to the IPv4 DHCP option. 238 Code Len Data 239 +------+------+------+------+------+-- --+-----+ 240 | code | len | URI ... | 241 +------+------+------+------+------+-- --+-----+ 243 o Code: The Captive-Portal DHCPv6Option (TBA2) 245 o Len: The length, in octets of the URI. 247 o URI: The URI of the authentication page that the user should 248 connect to. 250 4. The Captive-Portal IPv6 RA Option 252 This section describes the Captive-Portal Router Advertisement 253 option. 255 0 1 2 3 256 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 257 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 258 | Type | Length | URI . 259 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ . 260 . . 261 . . 262 . . 263 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 264 Figure 2: Captive-Portal RA Option Format 266 Type TBA3 268 Length 8-bit unsigned integer. The length of the option (including 269 the Type and Length fields) in units of 8 bytes. 271 URI The URI of the authentication page that the user should connect 272 to. For the reasons described above, the implementer might want 273 to use an IP address literal instead of a DNS name. This should 274 be padded with NULL (0x0) to make the total option length 275 (including the Type and Length fields) a multiple of 8 bytes. 277 5. Use of the Captive-Portal Option 279 [ED NOTE: This option provides notice to the OS / User applications 280 that there is a CP. Because of differences in UI design between 281 Operating Systems, the exact behaviour by OS and Applications is left 282 to the OS vendor/Application Developer.] 284 The purpose of the Captive-Portal Option is to inform the operating 285 system and applications that they are behind a captive portal type 286 device and will need to authenticate before getting network access 287 (and how to reach the authentication page). What is done with this 288 information is left up to the operating system and application 289 vendors. This document provides a very high level example of what 290 could be done with this information. 292 Many operating systems / applications already include a "connectivity 293 test" to determine if they are behind a captive portal (for example, 294 attempting to fetch a specific URL and looking for a specific string 295 (such as "Success"). These tests sometimes fail or take a long time 296 to determine when they are behind a CP, but are usually effective for 297 determining that the captive portal has been satisfied. These tests 298 will continue to be needed, because there is currently no definitive 299 signal from the captive portal that it has been satisfied. [ Editor 300 note: It may be useful to write another document that specifies how a 301 client can determine that it has passed the CP. This document could 302 also contain advice to implementors on only intercepting actually 303 needed ports, how to advertise that the CP needs to be satisfied 304 *again*, etc. This should not be done in this document though. ] The 305 connectivity test may also need to be used if the captive portal 306 times out the user session and needs the user to re-authenticate. 307 The operating system may still find the information about the captive 308 portal URI useful in this case. 310 When the device is informed that it is behind a captive portal it 311 should: 313 1. Not initiate new IP connections until the CP has been satisfied 314 (other than those to the captive portal browser session and 315 connectivity checks). Existing connections should be quiesced 316 (this will happen more often than some expect -- for example, the 317 user purchases 1 hour of Internet at a cafe and stays there for 3 318 hours -- this will "interrupt" the user a few times). 320 2. Present a dialog box to the user informing them that they are 321 behind a captive portal and ask if they wish to proceed. 323 3. If the user elects to proceed, the device should initiate a 324 connection to the captive portal login page using a web browser 325 configured with a separate cookie store, and without a proxy 326 server. If there is a VPN in place, this connection should be 327 made outside of the VPN and the user should be informed that 328 connection is outside the VPN. Some captive portals send the 329 user a cookie when they authenticate (so that the user can re- 330 authenticate more easily in the future) - the browser should keep 331 these CP cookies separate from other cookies. 333 4. Once the user has authenticated, normal IP connectivity should 334 resume. The CP success page should contain a string, e.g 335 "CP_SATISFIED." The OS can then use this string to provide 336 further information to the user. 338 5. The device should (using an OS dependent method) expose to the 339 user / user applications that they have connected though a 340 captive portal (for example by creating a file in /proc/net/ 341 containing the interface and captive portal URI). This should 342 continue until the network changes, or a new DHCP message without 343 the CP is received. 345 6. IANA Considerations 347 This document defines two DHCP Captive-Portal options, one for IPv6 348 and one for IPv6. It requires assignment of an option code (TBA1) to 349 be assigned from "Bootp and DHCP options" registry (http://www.iana 350 .org/assignments/ bootp-dhcp-parameters/bootp-dhcp-parameters.xml), 351 as specified in [RFC2939]. It also requires assignment of an option 352 code (TBA2) from the "DHCPv6 and DHCPv6 options" registry 353 (http://www.iana.org/assignments/dhcpv6-parameters/ 354 dhcpv6-parameters.xml). 356 IANA is also requested to assign an IPv6 RA Option Type code (TBA2) 357 from the "IPv6 Neighbor Discovery Option Formats" registry. Thanks 358 IANA! 360 7. Security Considerations 362 An attacker with the ability to inject DHCP messages could include 363 this option and so force users to contact an address of his choosing. 364 As an attacker with this capability could simply list himself as the 365 default gateway (and so intercept all the victim's traffic), this 366 does not provide them with significantly more capabilities. Fake 367 DHCP servers / fake RAs are currently a security concern - this 368 doesn't make them any better or worse. 370 Devices and systems that automatically connect to an open network 371 could potentially be tracked using the techniques described in this 372 document (forcing the user to continually authenticate, or exposing 373 their browser fingerprint.) However, similar tracking can already be 374 performed with the standard captive portal mechanisms, so this 375 technique does not give the attackers more capabilities. 377 By simplifying the interaction with the captive portal systems, and 378 doing away with the need for interception, we think that users will 379 be less likely to disable useful security safeguards like DNSSEC 380 validation, VPNs, etc. In addition, because the system knows that it 381 is behind a captive portal, it can know not to send cookies, 382 credentials, etc. Redirection to a portal where TLS can be used 383 without hijacking can ameliorate some of the implications of 384 connecting to a potentially malicious captive portal. 386 8. Acknowledgements 388 Thanks to Vint Cerf for the initial idea / asking me to write this. 389 Thanks to Wes George for supplying the IPv6 text. Thanks to Lorenzo 390 and Erik for the V6 RA kick in the pants. 392 Thanks to Fred Baker, Ted Lemon, Ole Troan and Asbjorn Tonnesen for 393 detailed review and comments. Also great thanks to Joel Jaeggli for 394 providing feedback and text. 396 9. Normative References 398 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 399 Requirement Levels", BCP 14, RFC 2119, March 1997. 401 Appendix A. Changes / Author Notes. 403 [RFC Editor: Please remove this section before publication ] 405 From 09 to 10: 407 o Ted Lemon and Joel Jaeggli: there's no benefit to insisting on an 408 ordering. I think you should just say that the ordering is 409 indeterminate, and if different mechanisms give non-equivalent 410 answers, this is likely to cause operational problems in practice. 412 From 08 to 09: 414 o Put back the DHCPv6 option, and made the fact that is separate 415 from the DHCPv4 option clearer (Ted Lemon) 417 From 07 to 08: 419 o Incorporated comments from Ted Lemon. Made the document much 420 shorter. 422 o Some cleanup. 424 From 06 to 07: 426 o Incoroprated a bunch of comments from Asbjorn Tonnesen 428 o Clarified that this document is only for the DHCP bits, not 429 everything. 431 o CP's *can* do HTTP redirects to DNS banes, as long as they allow 432 access to all needed services. 434 From 05 to 06: 436 o Integrated comments from Joel, as below 438 o Better introduction text, around the "kludgy hacks" section. 440 o Better "neither condones nor condems" text 442 o Fingerprint text. 444 o Some discussions on the v4 literal stuff. 446 o More Security Consideration text. 448 From 04 to 05: 450 o Integrated comments, primarily from Fred Baker. 452 From 03 to 04: 454 o Some text cleanup for readability. 456 o Some disclaimers about it working better on initial connection 457 versus CP timeout. 459 o Some more text explaining that CP interception is 460 indistinguishable from an attack. 462 o Connectivity Check test. 464 o Posting just before the draft cutoff - "I love deadlines. I love 465 the whooshing noise they make as they go by." -- Douglas Adams, 466 The Salmon of Doubt 468 From -02 to 03: 470 o Removed the DHCPv6 stuff (as suggested / requested by Erik Kline) 472 o Simplified / cleaned up text (I'm inclined to waffle on, then trim 473 the fluff) 475 o This was written on a United flight with in-flight WiFi - 476 unfortunately I couldn't use it because their CP was borked. :-P 478 From -01 to 02: 480 o Added the IPv6 RA stuff. 482 From -00 to -01: 484 o Many nits and editorial changes. 486 o Whole bunch of extra text and review from Wes George on v6. 488 From initial to -00. 490 o Nothing changed in the template! 492 Authors' Addresses 494 Warren Kumari 495 Google 496 1600 Amphitheatre Parkway 497 Mountain View, CA 94043 498 US 500 Email: warren@kumari.net 502 Olafur Gudmundsson 503 Shinkuro Inc. 504 4922 Fairmont Av, Suite 250 505 Bethesda, MD 20814 506 USA 508 Email: ogud@ogud.com 510 Paul Ebersman 511 Comcast 513 Email: ebersman-ietf@dragon.net 514 Steve Sheng 515 Internet Corporation for Assigned Names and Numbers 516 12025 Waterfront Drive, Suite 300 517 Los Angeles 90094 518 United States of America 520 Phone: +1.310.301.5800 521 Email: steve.sheng@icann.org