- From: Mark Nottingham <mnot@mnot.net>
- Date: Mon, 26 Jul 2021 12:43:44 +1000
- To: Joseph Salowey <joe@salowey.net>
- Cc: secdir@ietf.org, draft-ietf-httpbis-bcp56bis.all@ietf.org, ietf-http-wg@w3.org, last-call@ietf.org
Thanks for the review. I've opened an issue to track it: https://github.com/httpwg/http-extensions/issues/1582 Responses below. > On 26 Jul 2021, at 3:30 am, Joseph Salowey via Datatracker <noreply@ietf.org> wrote: > Major Issues: > > + I had trouble with section 4.12 Client Authentication which states: > > "Applications can use HTTP authentication Section 11 of > [I-D.ietf-httpbis-semantics] to identify clients. The Basic authentication > scheme [RFC7617] MUST NOT be used unless the underlying transport is > authenticated, integrity-protected and confidential (e.g., as provided the > "HTTPS" URI scheme, or another using TLS). The Digest scheme [RFC7616] MUST NOT > be used unless the underlying transport is similarly secure, or the chosen hash > algorithm is not "MD5"." > > I'm not sure what the "or chosen hash algorithm is not "MD5" is meant to say. > What I think the document should say is: > > The Digest scheme [RFC7616] MUST NOT be used unless the underlying transport is > similarly secure. The "MD5" digest algorithm MUST NOT be used. Hmm. I forgot that RFC7616 has a pre-existing requirement in Section 5.1: > If Digest Authentication is being used, it SHOULD be over a secure channel like HTTPS [RFC2818]. Are we saying that that SHOULD is really a MUST for all uses of HTTP, or just those in scope for this document? Likewise for the effective deprecation of md5. If so, perhaps the easiest thing to do would be to state that clearly; e.g., """ [RFC7616] Section 5.1 recommends that the Digest scheme be used over a secure channel like HTTPS. This document strengthens that recommendation to MUST, and deprecates the md5 hash algorithm in the Digest scheme. """ ... and listing 7616 as being updated by this document. Thoughts? > + There is a security consideration that I think the document should cover. > Many HTTP based protocols make heavy use of bearer tokens, such as session > cookies, for authentication and authorization purposes. This means that an > attacker that can eavesdrop on HTTP communications can often escalate their > privilege to perform operations on resources. I think you could add this to > the security considerations: > > " Section 4.4.2 requires support for 'https' URLs, and discourages the use of > 'http' URLs, to provide authentication, integrity and confidentiality, as well > as mitigate pervasive monitoring attacks. Many HTTP based protocols make heavy > use of bearer tokens, such as session cookies, for authentication and > authorization purposes. This means that an attacker that can eavesdrop on HTTP > communications can often escalate their privilege to perform operations on > resources. " See https://github.com/httpwg/http-extensions/commit/fe3f298e4d60514306e391398cc870abaedd8bf9 > Minor Issues: > > + Section 4.5.1 - This could be a good place to mention RFC-8470 on TLS 1.3 > early data which can also be a source of GET request replay See https://github.com/httpwg/http-extensions/commit/b03c376179e278deb3d994eac152c6a131702840 Cheers, -- Mark Nottingham https://www.mnot.net/
Received on Monday, 26 July 2021 02:44:04 UTC