Re: [PATCH v5] http: add http.sslVerifyStatus to check stapled OCSP responses
Grayson Gordon <[email protected]>
| Newsgroups | org.kernel.vger.git |
|---|---|
| Message-ID | <CALgUfNg1yryPygp_UVp9cGFfiUe7_6Uqx3ExBt=10Qh+PKG2QQ@mail.gmail.com> |
Junio, By "fail-closed" I was specifically referring to the case where no OCSP stapled response was provided. Failing open in this context would mean that, despite verifystatus being set, a response with no stapled response is ALLOWED. Maybe I'm just a very irregular human being lol. I'll update the adoc to be similar to what you provided. I refer to the edge cases/different behaviors that we talked about earlier in this thread with the older curl and gnutls versions in the commit message and keep the adoc as simple as possible. - Grayson On Tue, Aug 18, 2026 at 4:12 PM Junio C Hamano <[email protected]> wrote: > > graysongordon-gl <[email protected]> writes: > > > +http.sslVerifyStatus:: > > + Whether to check the revocation status of the server > > + certificate using the stapled OCSP response supplied during > > + the TLS handshake ("OCSP stapling"). Defaults to false. > > ++ > > +This is fail-closed: if the server staples no response, verification > > +fails. Set it per remote, e.g. > > +`http.https://example.com/.sslVerifyStatus`, rather than globally. > > I do not see us describe a knob or setting that can stop the > operation depending on some condition as "fail-closed". Can we > rephrase this for regular human beings? Perhaps > > Whether to refuse connecting to the server when its > certificate has been revoked. Default to false, allowing > connection even when its certificate is not known to be > still valid. > > or something like that might be a good starting point. After all, > the "check revocation and/or validity" is *not* the primary > objective from the end-user's point of view. Ensuring that they do > not talk to suspicious servers is. > > Thanks.