Opportunistic HTTP/2: Opportunistic Security for HTTP/2
A way to serve ordinary http addresses over an encrypted connection, without the client having to trust the server's identity.
Nottingham, M. and M. Thomson, "Opportunistic Security for HTTP/2", RFC 8164, DOI 10.17487/RFC8164, May 2017, <https://www.rfc-editor.org/info/rfc8164>.
@misc{rfc8164,
series = {Request for Comments},
number = 8164,
howpublished = {RFC 8164},
publisher = {RFC Editor},
doi = {10.17487/RFC8164},
url = {https://www.rfc-editor.org/info/rfc8164},
author = {M. Nottingham and M. Thomson},
title = {{Opportunistic Security for HTTP/2}},
pagetotal = 10,
year = 2017,
month = may,
} RFC text
Contents
Internet Engineering Task Force (IETF) M. Nottingham
Request for Comments: 8164
Category: Experimental M. Thomson
ISSN: 2070-1721 Mozilla
May 2017
Opportunistic Security for HTTP/2
Abstract
This document describes how "http" URIs can be accessed using
Transport Layer Security (TLS) and HTTP/2 to mitigate pervasive
monitoring attacks. This mechanism not a replacement for "https"
URIs; it is vulnerable to active attacks.
Status of This Memo
This document is not an Internet Standards Track specification; it is
published for examination, experimental implementation, and
evaluation.
This document defines an Experimental Protocol for the Internet
community. This document is a product of the Internet Engineering
Task Force (IETF). It represents the consensus of the IETF
community. It has received public review and has been approved for
publication by the Internet Engineering Steering Group (IESG). Not
all documents approved by the IESG are a candidate for any level of
Internet Standard; see Section 2 of RFC 7841.
Information about the current status of this document, any errata,
and how to provide feedback on it may be obtained at
http://www.rfc-editor.org/info/rfc8164.
Copyright Notice
Copyright (c) 2017 IETF Trust and the persons identified as the
document authors. All rights reserved.
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. Goals and Non-goals ........................................3
1.2. Notational Conventions .....................................3
2. Using HTTP URIs over TLS ........................................3
2.1. Alternative Server Opt-In ..................................4
2.2. Interaction with "https" URIs ..............................5
2.3. The "http-opportunistic" Well-Known URI ....................5
3. IANA Considerations .............................................6
4. Security Considerations .........................................7
4.1. Security Indicators ........................................7
4.2. Downgrade Attacks ..........................................7
4.3. Privacy Considerations .....................................7
4.4. Confusion regarding Request Scheme .........................7
4.5. Server Controls ............................................8
5. References ......................................................8
5.1. Normative References .......................................8
5.2. Informative References .....................................9
Acknowledgements ...................................................9
Authors' Addresses ................................................10
1. Introduction
This document describes a use of HTTP Alternative Services [RFC7838]
to decouple the URI scheme from the use and configuration of
underlying encryption. It allows an "http" URI [RFC7230] to be
accessed using HTTP/2 and Transport Layer Security (TLS) [RFC5246]
with Opportunistic Security [RFC7435].
This document describes a usage model whereby sites can serve "http"
URIs over TLS, thereby avoiding the problem of serving Mixed Content
(described in [W3C.CR-mixed-content-20160802]) while still providing
protection against passive attacks.
Opportunistic Security does not provide the same guarantees as using
TLS with "https" URIs, because it is vulnerable to active attacks,
and does not change the security context of the connection.
Normally, users will not be able to tell that it is in use (i.e.,
there will be no "lock icon").
1.1. Goals and Non-goals
The immediate goal is to make the use of HTTP more robust in the face
of pervasive passive monitoring [RFC7258].
A secondary (but significant) goal is to provide for ease of
implementation, deployment, and operation. This mechanism is
expected to have a minimal impact upon performance and require
trivial administrative effort to configure.
Preventing active attacks (such as man-in-the-middle attacks) is a
non-goal for this specification. Furthermore, this specification is
not intended to replace or offer an alternative to "https", since
"https" both prevents active attacks and invokes a more stringent
security model in most clients.
1.2. Notational Conventions
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. Using HTTP URIs over TLS
An origin server that supports the resolution of "http" URIs can
indicate support for this specification by providing an alternative
service advertisement [RFC7838] for a protocol identifier that uses
TLS, such as "h2" [RFC7540]. Such a protocol MUST include an
explicit indication of the scheme of the resource. This excludes
HTTP/1.1; HTTP/1.1 clients are forbidden from including the absolute
form of a URI in requests to origin servers (see Section 5.3.1 of
[RFC7230]).
A client that receives such an advertisement MAY make future requests
intended for the associated origin [RFC6454] to the identified
service (as specified by [RFC7838]), provided that the alternative
service opts in as described in Section 2.1.
A client that places the importance of protection against passive
attacks over performance might choose to withhold requests until an
encrypted connection is available. However, if such a connection
cannot be successfully established, the client can resume its use of
the cleartext connection.
A client can also explicitly probe for an alternative service
advertisement by sending a request that bears little or no sensitive
information, such as one with the OPTIONS method. Likewise, clients
with existing alternative services information could make such a
request before they expire, in order minimize the delays that might
be incurred.
Client certificates are not meaningful for URLs with the "http"
scheme; therefore, clients creating new TLS connections to
alternative services for the purposes of this specification MUST NOT
present them. A server that also provides "https" resources on the
same port can request a certificate during the TLS handshake, but it
MUST NOT abort the handshake if the client does not provide one.
2.1. Alternative Server Opt-In
For various reasons, it is possible that the server might become
confused about whether requests' URLs have an "http" or "https"
scheme (see Section 4.4). To ensure that the alternative service has
opted into serving "http" URLs over TLS, clients are required to
perform additional checks before directing "http" requests to it.
Clients MUST NOT send "http" requests over a secured connection,
unless the chosen alternative service presents a certificate that is
valid for the origin as defined in [RFC2818]. Using an authenticated
alternative service establishes "reasonable assurances" for the
purposes of [RFC7838]. In addition to authenticating the server, the
client MUST have obtained a valid "http-opportunistic" response for
an origin (as per Section 2.3) using the authenticated connection.
An exception to the latter restriction is made for requests for the
"http-opportunistic" well-known URI.
For example, assuming the following request is made over a TLS
connection that is successfully authenticated for those origins, the
following request/response pair would allow requests for the origins
"http://www.example.com" or "http://example.com" to be sent using a
secured connection:
HEADERS
+ END_STREAM
+ END_HEADERS
:method = GET
:scheme = http
:authority = example.com
:path = /.well-known/http-opportunistic
HEADERS
:status = 200
content-type = application/json
DATA
+ END_STREAM
[ "http://www.example.com", "http://example.com" ]
This document describes multiple origins, but only for operational
convenience. Only a request made to an origin (over an authenticated
connection) can be used to acquire the "http-opportunistic" resource
for that origin. Thus, in the example, the request to
"http://example.com" cannot be assumed to also provide a
representation of the "http-opportunistic" resource for
"http://www.example.com".
2.2. Interaction with "https" URIs
Clients MUST NOT send "http" and "https" requests on the same
connection. Similarly, clients MUST NOT send "http" requests for
multiple origins on the same connection.
2.3. The "http-opportunistic" Well-Known URI
This specification defines the "http-opportunistic" well-known URI
[RFC5785]. A client is said to have a valid "http-opportunistic"
response for a given origin when:
o The client has requested the well-known URI from the origin over
an authenticated connection and a 200 (OK) response was provided,
o That response is fresh [RFC7234] (potentially through revalidation
[RFC7232]),
o That response has the media type "application/json",
o That response's payload, when parsed as JSON [RFC7159], contains
an array as the root, and
o The array contains a string that is a case-insensitive, character-
for-character match for the origin in question, serialized into
Unicode as per Section 6.1 of [RFC6454].
A client MAY treat an "http-opportunistic" resource as invalid if
values it contains are not strings.
This document does not define semantics for "http-opportunistic"
resources on an "https" origin, nor does it define semantics if the
resource includes "https" origins.
Allowing clients to cache the "http-opportunistic" resource means
that all alternative services need to be able to respond to requests
for "http" resources. A client is permitted to use an alternative
service without acquiring the "http-opportunistic" resource from that
service.
A client MUST NOT use any cached copies of an "http-opportunistic"
resource that was acquired (or revalidated) over an unauthenticated
connection. To avoid potential errors, a client can request or
revalidate the "http-opportunistic" resource before using any
connection to an alternative service.
Clients that use cached "http-opportunistic" responses MUST ensure
that their cache is cleared of any responses that were acquired over
an unauthenticated connection. Revalidating an unauthenticated
response using an authenticated connection does not ensure the
integrity of the response.
3. IANA Considerations
This specification registers the following well-known URI [RFC5785]:
o URI Suffix: http-opportunistic
o Change Controller: IETF
o Specification Document(s): Section 2.3 of RFC 8164
o Related Information:
4. Security Considerations
4.1. Security Indicators
User agents MUST NOT provide any special security indicators when an
"http" resource is acquired using TLS. In particular, indicators
that might suggest the same level of security as "https" MUST NOT be
used (e.g., a "lock device").
4.2. Downgrade Attacks
A downgrade attack against the negotiation for TLS is possible.
For example, because the "Alt-Svc" header field [RFC7838] likely
appears in an unauthenticated and unencrypted channel, it is subject
to downgrade by network attackers. In its simplest form, an attacker
that wants the connection to remain in the clear need only strip the
"Alt-Svc" header field from responses.
4.3. Privacy Considerations
Cached alternative services can be used to track clients over time,
e.g., using a user-specific hostname. Clearing the cache reduces the
ability of servers to track clients; therefore, clients MUST clear
cached alternative service information when clearing other origin-
based state (i.e., cookies).
4.4. Confusion regarding Request Scheme
HTTP implementations and applications sometimes use ambient signals
to determine if a request is for an "https" resource; for example,
they might look for TLS on the stack or a server port number of 443.
This might be due to expected limitations in the protocol (the most
common HTTP/1.1 request form does not carry an explicit indication of
the URI scheme, and the resource might have been developed assuming
HTTP/1.1), or it may be because of how the server and application are
implemented (often, they are two separate entities, with a variety of
possible interfaces between them).
Any security decisions based upon this information could be misled by
the deployment of this specification, because it violates the
assumption that the use of TLS (or port 443) means that the client is
accessing an HTTPS URI and operating in the security context implied
by HTTPS.
Therefore, server implementers and administrators need to carefully
examine the use of such signals before deploying this specification.
4.5. Server Controls
This specification requires that a server send both an alternative
service advertisement and host content in a well-known location to
send HTTP requests over TLS. Servers SHOULD take suitable measures
to ensure that the content of the well-known resource remains under
their control. Likewise, because the "Alt-Svc" header field is used
to describe policies across an entire origin, servers SHOULD NOT
permit user content to set or modify the value of this header.
5. References
5.1. Normative References
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
Requirement Levels", BCP 14, RFC 2119,
DOI 10.17487/RFC2119, March 1997,
<http://www.rfc-editor.org/info/rfc2119>.
[RFC2818] Rescorla, E., "HTTP Over TLS", RFC 2818,
DOI 10.17487/RFC2818, May 2000,
<http://www.rfc-editor.org/info/rfc2818>.
[RFC5246] Dierks, T. and E. Rescorla, "The Transport Layer Security
(TLS) Protocol Version 1.2", RFC 5246,
DOI 10.17487/RFC5246, August 2008,
<http://www.rfc-editor.org/info/rfc5246>.
[RFC5785] Nottingham, M. and E. Hammer-Lahav, "Defining Well-Known
Uniform Resource Identifiers (URIs)", RFC 5785,
DOI 10.17487/RFC5785, April 2010,
<http://www.rfc-editor.org/info/rfc5785>.
[RFC6454] Barth, A., "The Web Origin Concept", RFC 6454,
DOI 10.17487/RFC6454, December 2011,
<http://www.rfc-editor.org/info/rfc6454>.
[RFC7159] Bray, T., Ed., "The JavaScript Object Notation (JSON) Data
Interchange Format", RFC 7159, DOI 10.17487/RFC7159, March
2014, <http://www.rfc-editor.org/info/rfc7159>.
[RFC7230] Fielding, R., Ed. and J. Reschke, Ed., "Hypertext Transfer
Protocol (HTTP/1.1): Message Syntax and Routing",
RFC 7230, DOI 10.17487/RFC7230, June 2014,
<http://www.rfc-editor.org/info/rfc7230>.
[RFC7232] Fielding, R., Ed. and J. Reschke, Ed., "Hypertext Transfer
Protocol (HTTP/1.1): Conditional Requests", RFC 7232,
DOI 10.17487/RFC7232, June 2014,
<http://www.rfc-editor.org/info/rfc7232>.
[RFC7234] Fielding, R., Ed., Nottingham, M., Ed., and J. Reschke,
Ed., "Hypertext Transfer Protocol (HTTP/1.1): Caching",
RFC 7234, DOI 10.17487/RFC7234, June 2014,
<http://www.rfc-editor.org/info/rfc7234>.
[RFC7540] Belshe, M., Peon, R., and M. Thomson, Ed., "Hypertext
Transfer Protocol Version 2 (HTTP/2)", RFC 7540,
DOI 10.17487/RFC7540, May 2015,
<http://www.rfc-editor.org/info/rfc7540>.
[RFC7838] Nottingham, M., McManus, P., and J. Reschke, "HTTP
Alternative Services", RFC 7838, DOI 10.17487/RFC7838,
April 2016, <http://www.rfc-editor.org/info/rfc7838>.
5.2. Informative References
[RFC7258] Farrell, S. and H. Tschofenig, "Pervasive Monitoring Is an
Attack", BCP 188, RFC 7258, DOI 10.17487/RFC7258, May
2014, <http://www.rfc-editor.org/info/rfc7258>.
[RFC7435] Dukhovni, V., "Opportunistic Security: Some Protection
Most of the Time", RFC 7435, DOI 10.17487/RFC7435,
December 2014, <http://www.rfc-editor.org/info/rfc7435>.
[W3C.CR-mixed-content-20160802]
West, M., "Mixed Content", World Wide Web Consortium CR
CR-mixed-content-20160802, August 2016,
<https://www.w3.org/TR/2016/CR-mixed-content-20160802>.
Acknowledgements
Mike Bishop contributed significant text to this document.
Thanks to Patrick McManus, Stefan Eissing, Eliot Lear, Stephen
Farrell, Guy Podjarny, Stephen Ludin, Erik Nygren, Paul Hoffman, Adam
Langley, Eric Rescorla, Julian Reschke, Kari Hurtta, and Richard
Barnes for their feedback and suggestions.
Authors' Addresses
Mark Nottingham
Email: mnot@mnot.net
URI: https://www.mnot.net/
Martin Thomson
Mozilla
Email: martin.thomson@gmail.com
ISO Plain Language
On this page
§1.1Goals and Non-Goals
The document describes a use of HTTP Alternative Services to decouple the URI scheme from the use and configuration of the underlying encryption. It lets an http URI be accessed using HTTP/2 and Transport Layer Security with Opportunistic Security.
The usage model is that sites can serve http URIs over TLS. That avoids the problem of serving mixed content while still giving protection against passive attacks. Opportunistic Security does not give the same guarantees as TLS with https URIs, because it is vulnerable to active attacks, and it does not change the security context of the connection. Normally, users will not be able to tell that it is in use: there is no lock icon.
The immediate goal is to make the use of HTTP more robust in the face of pervasive passive monitoring. The secondary, but significant, goal is ease of implementation, deployment and operation: the mechanism is expected to have a minimal impact on performance and to require trivial administrative effort to configure.
Two things are explicitly not goals. Preventing active attacks such as man-in-the-middle attacks is a non-goal. And the specification is not intended to replace or offer an alternative to https, because https both prevents active attacks and invokes a more stringent security model in most clients.
The aim is to make ordinary web traffic harder to watch on the network, and to do it without asking sites to change their addresses or do much work. What it explicitly will not do is stop someone in the middle of the connection from interfering.
- Resist passive monitoringThe immediate goal is to make the use of HTTP more robust in the face of pervasive passive monitoring.
- Easy to operateEase of implementation, deployment and operation, with minimal impact on performance and only trivial administrative effort.
- Avoid mixed contentServing http URIs over TLS avoids the problem of serving mixed content while still giving protection against passive attacks.
- Not preventing active attacksPreventing man-in-the-middle attacks is explicitly a non-goal of this specification.
- Not replacing httpsThe specification is not intended to replace or offer an alternative to https URIs.
§2Using HTTP URIs over TLS
An origin server that supports the resolution of http URIs can indicate support for this specification by providing an alternative service advertisement for a protocol identifier that uses TLS, such as h2. Such a protocol MUST include an explicit indication of the scheme of the resource, which excludes HTTP/1.1, because HTTP/1.1 clients are forbidden from including the absolute form of a URI in requests to origin servers.
A client that receives such an advertisement MAY make future requests intended for the associated origin to the identified service, provided that the alternative service opts in as described in Section 2.1. A client that places the importance of protection against passive attacks over performance might choose to withhold requests until an encrypted connection is available; if such a connection cannot be successfully established, the client can resume its use of the cleartext connection.
A client can also probe explicitly for an alternative service advertisement by sending a request that carries little or no sensitive information, such as one with the OPTIONS method. A client that already holds alternative services information could make such a request before that information expires, to reduce the delays that might otherwise be incurred.
Client certificates are not meaningful for URLs with the http scheme, so clients creating new TLS connections to alternative services for the purposes of this specification MUST NOT present them. A server that also provides https resources on the same port can request a certificate during the TLS handshake, but it MUST NOT abort the handshake if the client does not provide one.
The server says, in a header, that the same site can be reached over TLS at a given protocol. The client notes that and may use it for later requests. Client certificates are left out, because a certificate identifies the client and there is nothing here for it to identify.
- GET /.well-known/http-opportunisticSent over an authenticated TLS connection, with the scheme recorded as http.
- 200 OK, application/jsonThe body is a JSON array listing the origins the server is willing to serve over TLS.
- GET / over TLSThe client may now send http requests over the encrypted connection, because the certificate was valid for the origin and the well-known resource named it.
§2.3The Server Opt-In and the Well-Known URI
The server might become confused about whether a request's URL has an http or an https scheme, so clients must make additional checks before directing http requests to an alternative service. Clients MUST NOT send http requests over a secured connection unless the chosen alternative service presents a certificate that is valid for the origin. In addition to authenticating the server, the client MUST have obtained a valid http-opportunistic response for an origin using the authenticated connection. The exception is requests for the http-opportunistic well-known URI itself. The RFC gives the exchange: a GET for the well-known path with the scheme recorded as http, answered with a 200 and the media type application/json, whose body is an array of the origins the server will serve over TLS.
The specification describes multiple origins, but only for operational convenience. Only a request made to an origin, over an authenticated connection, can be used to acquire the http-opportunistic resource for that origin, so a request to one form of a name cannot be assumed to cover another form of it.
Clients MUST NOT send http and https requests on the same connection, and MUST NOT send http requests for multiple origins on the same connection.
The http-opportunistic well-known URI is a resource that a client has a valid response for when all of the following hold: the client requested the well-known URI from the origin over an authenticated connection and a 200 response was provided; that response is fresh, possibly through revalidation; it has the media type application/json; its payload, parsed as JSON, has an array as the root; and the array contains a string that matches the origin in question, case-insensitively and character for character, serialised into Unicode as described in RFC 6454. A client MAY treat the resource as invalid if the values in it are not strings.
Allowing clients to cache this resource means that all alternative services need to be able to respond to requests for http resources, and a client is permitted to use an alternative service without acquiring the resource from that service. A client MUST NOT use any cached copy of the resource that was acquired or revalidated over an unauthenticated connection, and a client that uses cached responses MUST ensure its cache is cleared of any response acquired over an unauthenticated connection. Revalidating an unauthenticated response over an authenticated connection does not make the response trustworthy.
Before sending ordinary http requests over the encrypted connection, the client checks two things: the certificate must be valid for the site being contacted, and the site must have listed that address in a standard file at a standard path. The listing is fetched over the encrypted connection, so a fake server cannot supply it.
HEADERS
+ END_STREAM
+ END_HEADERS
:method = GET
:scheme = http
:authority = example.com
:path = /.well-known/http-opportunistic
HEADERS
:status = 200
content-type = application/json
DATA
+ END_STREAM
[ "http://www.example.com", "http://example.com" ]§4Security Considerations
On security indicators, user agents MUST NOT provide any special security indicators when an http resource is acquired using TLS. In particular, indicators that might suggest the same level of security as https MUST NOT be used, such as a lock device.
On downgrade attacks, the RFC says one against the negotiation for TLS is possible. Because the Alt-Svc header field likely appears in an unauthenticated and unencrypted channel, it is subject to downgrade by network attackers. In its simplest form, an attacker that wants the connection to remain in the clear need only strip the Alt-Svc header field from responses.
On privacy, cached alternative services can be used to track clients over time, for example with a user-specific hostname. Clearing the cache reduces the ability of servers to track clients, so clients MUST clear cached alternative service information when they clear other origin-based state, such as cookies.
On confusion about the request scheme, HTTP implementations and applications sometimes use ambient signals to decide whether a request is for an https resource, for example by looking for TLS on the stack or for a server port number of 443. This may be because of a limitation in the protocol, since the most common HTTP/1.1 request form does not carry an explicit indication of the URI scheme, or because of how the server and application are implemented. Any security decision based on that information could be misled by the deployment of this specification, because it breaks the assumption that the use of TLS or port 443 means the client is accessing an https URI and operating in the security context that implies. Server implementers and administrators therefore need to examine the use of such signals carefully before deploying this specification.
On server controls, the specification requires a server to send both an alternative service advertisement and host content in a well-known location for http requests over TLS to be sent. Servers SHOULD take suitable measures to ensure that the content of the well-known resource remains under their control. And because the Alt-Svc header field describes policies across an entire origin, servers SHOULD NOT permit user content to set or modify its value.
The RFC is blunt about the limits. The browser must not show a lock icon, because the protection is weaker than https. An attacker on the network can remove the advertisement and leave the connection in the clear. And any server logic that decides whether a request is secure by looking for TLS or port 443 will get the wrong answer once this is deployed.
§3IANA Considerations
The specification registers one well-known URI with IANA. The URI suffix is http-opportunistic. The change controller is the IETF. The specification document is Section 2.3 of RFC 8164.
The RFC asks for one name to be reserved in the registry of well-known URI suffixes: http-opportunistic.