http J. Yasskin
Internet-Draft Google
Intended status: Standards Track August 31, 2017
Expires: March 4, 2018
Origin-signed HTTP Responses
draft-yasskin-http-origin-signed-responses-00
Abstract
This document explores how a server can send particular responses
that are authoritative for an origin, when the server itself is not
authoritative for that origin. For now, it focuses on the
constraints covering any such mechanism.
Note to Readers
Discussion of this draft takes place on the HTTP working group
mailing list (ietf-http-wg@w3.org), which is archived at
https://lists.w3.org/Archives/Public/ietf-http-wg/.
The source code and issues list for this draft can be found in
https://github.com/WICG/webpackage.
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 March 4, 2018.
Copyright Notice
Copyright (c) 2017 IETF Trust and the persons identified as the
document authors. All rights reserved.
Yasskin Expires March 4, 2018 [Page 1]
Internet-Draft Origin-signed HTTP Responses August 2017
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
2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 3
3. Use cases . . . . . . . . . . . . . . . . . . . . . . . . . . 3
3.1. PUSHed subresources . . . . . . . . . . . . . . . . . . . 3
3.2. Explicit use of a CDN for subresources . . . . . . . . . 3
3.3. Subresource Integrity . . . . . . . . . . . . . . . . . . 4
3.4. Offline websites . . . . . . . . . . . . . . . . . . . . 4
4. Requirements and open questions . . . . . . . . . . . . . . . 5
4.1. Proof of origin . . . . . . . . . . . . . . . . . . . . . 5
4.1.1. The certificate and its chain . . . . . . . . . . . . 5
4.2. How much to sign . . . . . . . . . . . . . . . . . . . . 5
4.3. Response lifespan . . . . . . . . . . . . . . . . . . . . 6
4.3.1. Certificate revocation . . . . . . . . . . . . . . . 6
4.3.2. Response downgrade attacks . . . . . . . . . . . . . 7
4.4. Conveying the signed headers . . . . . . . . . . . . . . 7
5. Straw proposal . . . . . . . . . . . . . . . . . . . . . . . 8
6. Security considerations . . . . . . . . . . . . . . . . . . . 8
7. Privacy considerations . . . . . . . . . . . . . . . . . . . 8
8. IANA considerations . . . . . . . . . . . . . . . . . . . . . 9
9. References . . . . . . . . . . . . . . . . . . . . . . . . . 9
9.1. Normative References . . . . . . . . . . . . . . . . . . 9
9.2. Informative References . . . . . . . . . . . . . . . . . 9
9.3. URIs . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Appendix A. Acknowledgements . . . . . . . . . . . . . . . . . . 11
Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 11
1. Introduction
When I presented Web Packaging to DISPATCH [3], folks thought it
would make sense to split it into a way to sign individual HTTP
responses as coming from a particular origin, and separately a way to
bundle a collection of HTTP responses. This document explores the
constraints on any method of signing HTTP responses and briefly
sketches a possible solution to the constraints.
Yasskin Expires March 4, 2018 [Page 2]
Internet-Draft Origin-signed HTTP Responses August 2017
2. Terminology
Author The entity that controls the server for a particular origin
[RFC6454]. The author can get a CA to issue certificates for
their private keys and can run a TLS server for their origin.
Intermediate An entity that fetches signed HTTP responses from an
author or another intermediate and forwards them to another
intermediate or a client.
Client An entity that uses a signed HTTP response and needs to be
able to prove that the author vouched for it as coming from its
claimed origin.
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and
"OPTIONAL" in this document are to be interpreted as described in BCP
14 [RFC2119] [RFC8174] when, and only when, they appear in all
capitals, as shown here.
3. Use cases
3.1. PUSHed subresources
To reduce round trips, a server might use HTTP/2 PUSH to inject a
subresource from another server into the client's cache. If anything
about the subresource is expired or can't be verified, the client
would fetch it from the original server.
For example, if "https://example.com/index.html" includes
<script src="https://jquery.com/jquery-1.2.3.min.js">
Then to avoid the need to look up and connect to "jquery.com" in the
critical path, "example.com" might PUSH that resource ([RFC7540],
section 8.2), signed by "jquery.com".
3.2. Explicit use of a CDN for subresources
In order to speed up loading but still maintain control over its
content, an HTML page in a particular origin "O.com" could tell
clients to load its subresources from an intermediate CDN that's not
authoritative, but require that those resources be signed by "O.com"
so that the CDN couldn't modify the resources.
<img logicalsrc="https://O.com/img.png"
physicalsrc="https://cdn.com/O.com/img.png">
Yasskin Expires March 4, 2018 [Page 3]
Internet-Draft Origin-signed HTTP Responses August 2017
To make it easier to configure the right CDN for a given request,
computation of the "physicalsrc" could be encapsulated in a custom
element:
<cdn-img src="https://O.com/img.png"></cdn-img>
where the "<cdn-img>" implementation generates an appropriate "<img>"
based on, for example, a "<meta name="cdn-base">" tag elsewhere in
the page.
This could be used for some of the same purposes as SRI
(Section 3.3).
3.3. Subresource Integrity
The W3C WebAppSec group is investigating using signatures [4] in
[SRI]. They need a way to transmit the signature with the response,
which this proposal could provide.
However, their needs also differ in some significant ways:
1. The "integrity="ed25519-[public-key]"" attribute and CSP-based
ways of expressing a public key don't need the signing key to be
also trusted to sign arbitrary content for an origin.
2. Some uses of SRI want to constrain subresources to be vouched for
by a third-party, rather than just being signed by the
subresource's author.
While we can design this system to cover both origin-trusted and
simple-key signatures, we should check that this is better than
having two separate systems for the two kinds of signatures.
Note that while the current proposal for SRI describes signing only
the content of a resource, they may need to sign its name as well, to
prevent security vulnerabilities [5]. The details of what they need
to sign will affect whether and how they can use this proposal.
3.4. Offline websites
See https://github.com/WICG/webpackage and
[I-D.yasskin-dispatch-web-packaging]. This use requires origin-
signed resources to be bundled.
Yasskin Expires March 4, 2018 [Page 4]
Internet-Draft Origin-signed HTTP Responses August 2017
4. Requirements and open questions
4.1. Proof of origin
To verify that a thing came from a particular origin, for use in the
same context as a TLS connection, we need someone to vouch for the
signing key with as much verification as the signing keys used in
TLS. The obvious way to do this is to re-use the web PKI and CA
ecosystem.
We'll improve adoption if we can re-use existing TLS server
certificates, but if the risks of doing that are too high, we could
define a new Extended Key Usage ([RFC5280], section 4.2.1.12) that
requires CAs to issue new keys for this purpose. The rest of this
document will assume we can re-use existing TLS server certificates.
It's essential that an attacker who can convince the server to sign
some resource, cannot cause those signed bytes to be interpreted as
something else. As a result, signatures here need to use the same
format as TLS's signatures, specified in [I-D.ietf-tls-tls13] section
4.4.3, with a context string that's specific to this use.
4.1.1. The certificate and its chain
The client needs to be able to find the certificate vouching for the
signing key and a chain from that certificate to a trusted root. One
approach would be to include the certificate and its chain in the
signature metadata itself, but this wastes bytes when the same
certificate is used for multiple HTTP responses. If we decide to put
the signature in an HTTP header, certificates are also unusually
large for that context.
Another option is to pass a URL that the client can fetch to retrieve
the certificate and chain. To avoid extra round trips in fetching
that URL, it could be bundled (Section 3.4) with the signed content
or PUSHed (Section 3.1) with it.
4.2. How much to sign
The previous [I-D.thomson-http-content-signature] and
[I-D.burke-content-signature] schemes signed just the content, while
([I-D.cavage-http-signatures] could also sign the response headers
and the request method and path. However, the same path, response
headers, and content may mean something very different when retrieved
from a different server, so this document expects to include the
whole URL in the signed data as well.
Yasskin Expires March 4, 2018 [Page 5]
Internet-Draft Origin-signed HTTP Responses August 2017
The question of whether to include other request headers--primarily
the "accept*" family--is still open. These headers need to be
represented so that clients wanting a different language, say, can
avoid using the wrong-language response, but it's not obvious that
there's a security vulnerability if an attacker can spoof them. That
said, it's always safer to include everything in the signature.
In order to allow multiple clients to consume the same signed
response, the response shouldn't include the exact headers that any
particular client sends. For example, a Japanese resource wouldn't
include
accept-language: ja-JP, ja;q=0.9, en;q=0.8, zh;q=0.7, *;q=0.5
Instead, it would probably include just
accept-language: ja-JP, ja
and clients would use the same matching logic as for PUSH_PROMISE [7]
frame headers.
The rest of this document will assume that all request headers are
included in the signature.
4.3. Response lifespan
A normal HTTPS response is authoritative only for one client, for as
long as its cache headers say it should live. A signed response can
be re-used for many clients, and if it was generated while a server
was compromised, it can continue compromising clients even if their
requests happen after the server recovers. This signing scheme needs
to mitigate that risk.
4.3.1. Certificate revocation
Certificates are mis-issued and private keys are stolen, and in
response clients need to be able to stop trusting these certificates
as promptly as possible. Online revocation checks don't work [8], so
the industry has moved to pushed revocation lists and stapled OCSP
responses [RFC6066].
Pushed revocation lists work as-is to block trust in the certificate
signing a response, but the signatures need an explicit strategy to
staple OCSP responses. One option is to extend the certificate
download (Section 4.1.1) to include the OCSP response too, perhaps in
the TLS 1.3 CertificateEntry [9] format.
Yasskin Expires March 4, 2018 [Page 6]
Internet-Draft Origin-signed HTTP Responses August 2017
4.3.2. Response downgrade attacks
The signed content in a response might be vulnerable to attacks, such
as XSS, or might simply be discovered to be incorrect after
publication. Once the author fixes those vulnerabilities or
mistakes, clients should stop trusting the old signed content in a
reasonable amount of time. Similar to certificate revocation, I
expect the best option to be stapled "this version is still valid"
assertions with short expiration times.
These assertions could be structured as:
1. A signed minimum version number or timestamp for a set of request
headers: This requires that signed responses need to include a
version number or timestamp, but allows a server to provide a
single signature covering all valid versions.
2. A replacement for the whole response's signature. This requires
the author to separately re-sign each valid version and requires
each version to include a strong validator [RFC7232], but allows
intermediates to serve less data.
3. A replacement for the response's signature and an update for the
embedded "expires" and related cache-control HTTP headers
[RFC7234]. This naturally extends authors' intuitions about
cache expiration and the existing cache revalidation behavior to
signed responses. However, it also requires that the update
procedure for the response headers on the client must produce
something that's bytewise identical to the updated headers on the
server.
The signature also needs to include instructions to intermediates for
how to fetch updated validity assertions.
4.4. Conveying the signed headers
HTTP headers are traditionally munged by proxies, making it
impossible to guarantee that the client will see the same sequence of
bytes as the author wrote. In the HTTPS world, we have more end-to-
end header integrity, but it's still likely that there are enough
TLS-terminating proxies that the author's signatures would tend to
break before getting to the client.
Since proxies don't modify unknown content types, I expect to propose
an "application/http2" format to serialize the request headers,
response headers, and body so they can be signed. This could be as
simple as a series of HTTP/2 frames, or could
Yasskin Expires March 4, 2018 [Page 7]
Internet-Draft Origin-signed HTTP Responses August 2017
1. Allow longer contiguous bodies than HTTP/2's 16MB frame limit
[10], and
2. Use better compression than [RFC7541] for the non-confidential
headers.
To help the PUSHed subresources use case (Section 3.1), we might also
want to extend the "PUSH_PROMISE" frame type to include a signature,
and then there might be some way to include the request headers
directly in that frame.
5. Straw proposal
TBD
6. Security considerations
Authors MUST NOT include confidential information in a signed
response that an untrusted intermediate could forward, since the
response is only signed and not encrypted. Intermediates can read
the content.
Relaxing the requirement to consult DNS when determining authority
for an origin means that an attacker who possesses a valid
certificate no longer needs to be on-path to redirect traffic to
them; instead of modifying DNS, they need only convince the user to
visit another Web site in order to serve responses signed as the
target. This consideration and mitigations for it are shared by
[I-D.ietf-httpbis-origin-frame].
Signing a bad response can affect more users than simply serving a
bad response, since a served response will only affect users who make
a request while the bad version is live, while an attacker can
forward a signed response until its signature expires. Authors
should consider shorter signature expiration times than they use for
cache expiration times.
7. Privacy considerations
Normally, when a client fetches "https://o1.com/resource.js",
"o1.com" learns that the client is interested in the resource. If
"o1.com" signs "resource.js", "o2.com" serves it as "https://o2.com/
o1resource.js", and the client fetches it from there, then "o2.com"
learns that the client is interested, and if the client executes the
Javascript, that could also report the client's interest back to
"o1.com".
Yasskin Expires March 4, 2018 [Page 8]
Internet-Draft Origin-signed HTTP Responses August 2017
Often, "o2.com" already knew about the client's interest, because
it's the entity that directed the client to "o1resource.js", but
there may be cases where this leaks extra information.
For non-executable resource types, a signed response can improve the
privacy situation by hiding the client's interest from the original
author.
8. IANA considerations
This document has no actions for IANA.
9. References
9.1. 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, <https://www.rfc-
editor.org/info/rfc2119>.
[RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC
2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174,
May 2017, <https://www.rfc-editor.org/info/rfc8174>.
9.2. Informative References
[I-D.burke-content-signature]
Burke, B., "HTTP Header for digital signatures", draft-
burke-content-signature-00 (work in progress), March 2011.
[I-D.cavage-http-signatures]
Cavage, M. and M. Sporny, "Signing HTTP Messages", draft-
cavage-http-signatures-07 (work in progress), July 2017.
[I-D.ietf-httpbis-origin-frame]
Nottingham, M. and E. Nygren, "The ORIGIN HTTP/2 Frame",
draft-ietf-httpbis-origin-frame-04 (work in progress),
August 2017.
[I-D.ietf-tls-tls13]
Rescorla, E., "The Transport Layer Security (TLS) Protocol
Version 1.3", draft-ietf-tls-tls13-21 (work in progress),
July 2017.
Yasskin Expires March 4, 2018 [Page 9]
Internet-Draft Origin-signed HTTP Responses August 2017
[I-D.thomson-http-content-signature]
Thomson, M., "Content-Signature Header Field for HTTP",
draft-thomson-http-content-signature-00 (work in
progress), July 2015.
[I-D.yasskin-dispatch-web-packaging]
Yasskin, J., "Web Packaging", draft-yasskin-dispatch-web-
packaging-00 (work in progress), June 2017.
[RFC5280] Cooper, D., Santesson, S., Farrell, S., Boeyen, S.,
Housley, R., and W. Polk, "Internet X.509 Public Key
Infrastructure Certificate and Certificate Revocation List
(CRL) Profile", RFC 5280, DOI 10.17487/RFC5280, May 2008,
<https://www.rfc-editor.org/info/rfc5280>.
[RFC6066] Eastlake 3rd, D., "Transport Layer Security (TLS)
Extensions: Extension Definitions", RFC 6066,
DOI 10.17487/RFC6066, January 2011, <https://www.rfc-
editor.org/info/rfc6066>.
[RFC6454] Barth, A., "The Web Origin Concept", RFC 6454,
DOI 10.17487/RFC6454, December 2011, <https://www.rfc-
editor.org/info/rfc6454>.
[RFC7232] Fielding, R., Ed. and J. Reschke, Ed., "Hypertext Transfer
Protocol (HTTP/1.1): Conditional Requests", RFC 7232,
DOI 10.17487/RFC7232, June 2014, <https://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,
<https://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, <https://www.rfc-
editor.org/info/rfc7540>.
[RFC7541] Peon, R. and H. Ruellan, "HPACK: Header Compression for
HTTP/2", RFC 7541, DOI 10.17487/RFC7541, May 2015,
<https://www.rfc-editor.org/info/rfc7541>.
[SRI] Akhawe, D., Braun, F., Marier, F., and J. Weinberger,
"Subresource Integrity", World Wide Web Consortium
Recommendation REC-SRI-20160623, June 2016,
<http://www.w3.org/TR/2016/REC-SRI-20160623>.
Yasskin Expires March 4, 2018 [Page 10]
Internet-Draft Origin-signed HTTP Responses August 2017
9.3. URIs
[1] https://datatracker.ietf.org/doc/minutes-99-dispatch/
[2] https://github.com/mikewest/signature-based-sri
[3] https://github.com/mikewest/signature-based-sri/issues/5
[5] https://tools.ietf.org/html/rfc7540#section-8.2
[6] https://www.imperialviolet.org/2012/02/05/crlsets.html
[7] https://tlswg.github.io/tls13-spec/draft-ietf-tls-
tls13.html#ocsp-and-sct
[8] https://tools.ietf.org/html/rfc7540#section-4.2
Appendix A. Acknowledgements
Author's Address
Jeffrey Yasskin
Google
Email: jyasskin@chromium.org
Yasskin Expires March 4, 2018 [Page 11]