Network Working Group W. Kumari Internet-Draft Google Intended status: Informational O. Gudmundsson Expires: October 18, 2014 Shinkuro Inc. P. Ebersman Infoblox S. Sheng ICANN April 16, 2014 Captive-Portal identification in DHCP / RA draft-wkumari-dhc-capport-02 Abstract In many environments (such as hotels, coffee shops and other establishments that offer Internet service to customers), it is common to start new connections in a captive portal mode, i.e. highly restrict what the customer can do until the customer has accepted terms of service, provided payment information or authenticated. This document describes a DHCP option (and an RA extension) to inform clients that they are behind some sort of captive portal device, and that they will need to authenticate to get Internet Access. Status of This Memo This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet- Drafts is at http://datatracker.ietf.org/drafts/current/. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress." This Internet-Draft will expire on October 18, 2014. Copyright Notice Copyright (c) 2014 IETF Trust and the persons identified as the document authors. All rights reserved. Kumari, et al. Expires October 18, 2014 [Page 1] Internet-Draft DHCP Captive-Portal April 2014 This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (http://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License. Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 1.1. Requirements notation . . . . . . . . . . . . . . . . . . 3 2. Background . . . . . . . . . . . . . . . . . . . . . . . . . 3 2.1. DNS Redirection . . . . . . . . . . . . . . . . . . . . . 3 2.2. HTTP Redirection . . . . . . . . . . . . . . . . . . . . 4 2.3. IP Hijacking . . . . . . . . . . . . . . . . . . . . . . 4 3. The Captive-Portal DHCP Option . . . . . . . . . . . . . . . 4 4. The Captive-Portal RA Option . . . . . . . . . . . . . . . . 5 5. Use of the Captive-Portal Option . . . . . . . . . . . . . . 6 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 8 7. Security Considerations . . . . . . . . . . . . . . . . . . . 8 8. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 8 9. References . . . . . . . . . . . . . . . . . . . . . . . . . 9 9.1. Normative References . . . . . . . . . . . . . . . . . . 9 9.2. Informative References . . . . . . . . . . . . . . . . . 9 Appendix A. Changes / Author Notes. . . . . . . . . . . . . . . 9 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 9 1. Introduction In many environments (coffee shops and hotels), users need to connect to a captive portal device and agree to an acceptable use policy or provide billing information before they can access the Internet. In order to present the user with the captive portal web page, many devices perform DNS and / or HTTP and / or IP hijacks. As well as being kludgy hacks, these techniques looks very similar to attacks that DNSSEC and TLS protect against. This document describes a DHCP option (Captive-Portal) that informs DHCP clients that they are behind a captive portal device, and how to contact it. This document neither condones nor condemns captive portals; instead it recognises that they are here to stay, and attempts to improve the user's experience. Kumari, et al. Expires October 18, 2014 [Page 2] Internet-Draft DHCP Captive-Portal April 2014 1.1. Requirements notation The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in [RFC2119]. 2. Background Many Internet Service Providers (ISPs) that offer public Internet access require the user to first accept an Acceptable Use Policy (AUP) and / or provides billing information (such as their last name and / or room number in a hotel, credit card information, etc) through a web interface. In order to meet this requirement, some ISPs implement a captive portal (CP) - a system that intercepts user requests and redirects them to an interstitial login page. Captive portals intercept and redirects user requests in a number of ways, including: o DNS Redirection o IP Redirection o HTTP Redirection o Restricted scope addresses o Traffic blocking (until the user is authenticated) In order to ensure that the user is unable to access the Internet, captive portals usually implement IP based filters, or place the user in to a restricted VLAN or restricted IP range until after they have been authorized. 2.1. DNS Redirection The CP either intercepts all DNS traffic or advertises itself (for example using DHCP) as the recursive server for the network. Until the user has authenticated to the captive portal, the CP responds to all DNS requests listing the address of its web portal. Once the user has authenticated the CP returns the "correct" addresses. This technique has many shortcomings. It fails if the client is performing DNSSEC validation, or if the client already has the DNS information cached. Kumari, et al. Expires October 18, 2014 [Page 3] Internet-Draft DHCP Captive-Portal April 2014 2.2. HTTP Redirection In this implementation, the CP acts like a transparent HTTP proxy; but when it sees an HTTP request from an unauthenticated client, it intercepts the request and responds with an HTTP status code 302 to redirect the client to the Captive Portal Login. The issues with this technique include: o It fails if the user is only using HTTPS o It exposes various private user information, such as HTTP Cookies, etc. o It doesn't work if the client has a VPN and / or proxies their web traffic to an external web proxy. 2.3. IP Hijacking In this scenario, the captive portal intercepts connections to any IP address. It spoofs the destination IP address and "pretends" to be whatever the user tried to access. This technique has similar issues as the HTTP solution, but may also break other protocols, and may expose more of the users private information, etc. 3. The Captive-Portal DHCP Option The Captive Portal DHCP Option (TBA1) informs the DHCP client that it is behind a captive portal and provides the URI to access the authentication page. This is primarily intended to improve the user experiance; for the forseeable future captive portals will still need to implement the interception techniques to serve legacy clinets. This draft is not intended to provide guidance on how to implement a captive portal. As such, it assumes that the captive portal on a dual-stack or IPv6-only network is already capable of intercepting IPv6 traffic. However, in order to support IPv6 with the proposed DHCP option, there are some additional considerations. In a dual- stack network, the network supports both IPv4 and IPv6 protocols simultaneously, but can have a mix of IPv4-only, IPv6-only, and dual- stack devices using the network, meaning that it may be necessary to have parallel notifications via DHCPv4 and DHCPv6. IPv4-only and dual-stack devices can technically both support receiving the option via DHCPv4, but dual-stack implementations would need to ensure that the correct action would be taken for both IPv4 Kumari, et al. Expires October 18, 2014 [Page 4] Internet-Draft DHCP Captive-Portal April 2014 and IPv6 traffic despite only receiving an option via IPv4. For devices/networks that only speak IPv6, and to avoid this dependency on the implementation, a DHCPv6 option is necessary. [ED NOTE:] This is complicated by the fact that not all devices support DHCPv6, and thus it may be necessary to investigate other methods to notify IPv6-only devices of a captive portal. Since this option is only intended to help clients gracefully deal with networks that have a captive portal, it may be acceptable to note that if a client does not support DHCPv6, it simply won't be able to take advantage of this optimization, but will otherwise function normally. [/note] The format of the DHCP Captive-Portal DHCP option is identical for both DHCPv4 and DHCPv6 and is shown below. Code Len Data +------+------+------+------+------+-- --+-----+ | code | len | URI ... | +------+------+------+------+------+-- --+-----+ o Code: The Captive-Portal DHCP Option (TBA1 for DHCPv4, TBA2 for DHCPv6) o Len: The length, in octets of the URI. o URI: The URI of the authentication page that the user should connect to. The URI MUST be a URL with an IP-literal for the host portion (to remove the need to allow DNS from unauthenticated clients). The DHCPv4 URI MUST contain an IPv4 address, and the DHCPv6 URI MUST contain an IPv6 address (to account for IPv4 only or IPv6 only capable devices - not everyting is dual stack!) [ED NOTE: Using an address literal is less than ideal, but better than the alternatives. Recommending a DNS name means that the CP would need to allow DNS from unauthenticated clients (as we don't want to force users to use the CP's provided DNS) and some folk would use this to DNS Tunnel out. This would make the CP admin block external recursives).] 4. The Captive-Portal RA Option [ Ed: I'm far from an RA expert, but it was suggested that we shold advertize this via RA as well as DHCP. I think there are only 8 bits for Type, is it worth burning an option code on this? I have also specified that the option length should padded to multiples of 8 byte Kumari, et al. Expires October 18, 2014 [Page 5] Internet-Draft DHCP Captive-Portal April 2014 to better align with the examples I've seen (and I'm on a plane so cannot easily search for more!) Is this required / preferred, or is smaller RAs better? ] This section describes the Captive-Portal Router Advertisement option. 0 1 2 3 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 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type | Length | URI . +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ . . . . . . . +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Figure 2: Captive-Portal RA Option Format Type TBA3 Length 8-bit unsigned integer. The length of the option (including the Type and Length fields) in units of 8 bytes. URI The URI of the authentication page that the user should connect to. This should be padded with NULL (0x0) to make the total option length (including the Type and Length fields) a mutliple of 8 bytes. 5. Use of the Captive-Portal Option [ED NOTE: This section is, and probably will remain, fairly hand wavy. This option provides notice to the OS / User applications that there is a CP, but I think that the UI / etc is best designed / handled by the Operating System vendors / Application developers. ] The purpose of the Captive-Portal Option is to inform the operating system and applications that they are behind a captive portal type device and will need to authenticate before getting network access (and how to reach the authentication page). The Captive-Portal Option is defined for IPv6 RAs and IPv6 DHCP and IPv4 DHCP. The URIs in these SHOULD all be the same, but if they are not, the precedence is as follows: 1. IPv6 DHCP (most preferred) 2. IPv4 DHCP Kumari, et al. Expires October 18, 2014 [Page 6] Internet-Draft DHCP Captive-Portal April 2014 3. IPv6 RA (least preferred) This preference was chosen because an IPv6 capable client machine will first get an IP address (and possibly the Captive-Portal option) via RA, and will then get additional information via DHCP v6. The client is not fully configured until it has completed the DHCP step, and so this is "newer" information. The DHCP v4 preference is in the middle, because it is likely that this will be delpoyed first on many captive portals. Once IPv6 is deployed, we don't want legacy (forgotten!) configuration to override the "newer" configuration information. [Ed: This ordering is somewhat arbritary, as is the justification, but there should be *some* standard and this seemed as good as any! ] The exact method that the interaction with the user occurs is device / operating system / application dependent. The below is simply one option. When the device receives a DHCP response with the Captive-Portal Option it SHOULD: 1. Not initiate new IP connections until the CP has been satisfied. Existing connections should be quiesced. This will happen more often than some expect -- you buy 1h of Internet at a cafe and stay there for 3h -- this will "interrupt" you a few times). 2. Present a dialog box to the user informing that they are behind a captive portal and ask if they wish to proceed. 3. If the user elects to proceed, the device should initiate a connection to the captive portal login page using a web browser configured with a separate cookie store. Some captive portals send the user a cookie when they authenticate (so that the user can re-authenticate more easily in the future - the browser should keep these CP cookies separate from other cookies. 4. Once the user has authenticated (how does it know? HTTP response?! Probe (ugh?)) normal IP connectivity should resume. 5. The device should (using an OS dependent method) expose to the user / user applications that they have connected though a captive portal (for example by creating a file in /proc/net/ containing the interface and captive portal URI). This should continue until the network changes, or a new DHCP message without the CP is received. Kumari, et al. Expires October 18, 2014 [Page 7] Internet-Draft DHCP Captive-Portal April 2014 6. IANA Considerations This document defines DHCPv4 Captive-Portal option which requires assignment of DHCPv4 option code TBA1 assigned from "Bootp and DHCP options" registry (http://www.iana.org/assignments/ bootp-dhcp- parameters/bootp-dhcp-parameters.xml), as specified in [RFC2939]. The IANA is requested to also assign an option codes for the DHCPv6 Captive-Portal (TBA2) option from the "DHCPv6 and DHCPv6 options" registry (http:// www.iana.org/assignments/dhcpv6-parameters/ dhcpv6-parameters.xml). The IANA is also requested at assign an IPv6 RA Type code (TBA3) from the [TODO] registry. Thanks IANA! 7. Security Considerations An attacker with the ability to inject DHCP messages could include this option and so force users to contact him. As an attacker with this capability could simply list himself as the default gateway (and so see all the victim's traffic), we do not think this gives them significantly more capabilities. Fake DHCP servers / fake RAs are currently a security concern - this doesn't make them any better or worse. Devices and systems that automatically connect to open network could potentially be tracked using the techniques described in this document (forcing the user to continually authenticate, or exposing their browser fingerprint), but similar tracking could already be performed with the standard captive portal mechanisms, so this doesn't seem to give the attackers more capabilities. By simplifying the interaction with the captive portal systems, and doing away with the need for interception, we think that users will be less likely to disable useful security safeguards like DNSSEC validation, VPNs, etc. 8. Acknowledgements The primary author has discussed this idea with a number of folk, and asked them to assist by becoming co-authors. Unfortunately he has forgotten who many of them were; if you were one of them, I apologize. Thanks to Vint Cerf for the initial idea / asking me to write this. Thanks to Wes George for supplying the v6 text. Kumari, et al. Expires October 18, 2014 [Page 8] Internet-Draft DHCP Captive-Portal April 2014 9. References 9.1. Normative References [IANA.AS_Numbers] IANA, "Autonomous System (AS) Numbers", . [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, March 1997. 9.2. Informative References [I-D.ietf-sidr-iana-objects] Manderson, T., Vegoda, L., and S. Kent, "RPKI Objects issued by IANA", draft-ietf-sidr-iana-objects-03 (work in progress), May 2011. Appendix A. Changes / Author Notes. [RFC Editor: Please remove this section before publication ] From -01 to 02: o Added the IPv6 RA stuff. From -00 to -01: o Many nits and editorial changes. o Whole bunch of extra text and review from Wes George on v6. From initial to -00. o Nothing changed in the template! Authors' Addresses Warren Kumari Google 1600 Amphitheatre Parkway Mountain View, CA 94043 US Email: warren@kumari.net Kumari, et al. Expires October 18, 2014 [Page 9] Internet-Draft DHCP Captive-Portal April 2014 Olafur Gudmundsson Shinkuro Inc. 4922 Fairmont Av, Suite 250 Bethesda, MD 20814 USA Email: ogud@ogud.com Paul Ebersman Infoblox Email: ebersman-ietf@dragon.net Steve Sheng Internet Corporation for Assigned Names and Numbers 12025 Waterfront Drive, Suite 300 Los Angeles 90094 United States of America Phone: +1.310.301.5800 Email: steve.sheng@icann.org Kumari, et al. Expires October 18, 2014 [Page 10]