[0/2] git+curl 8.7.0 workaround

Message

Jeff King April 2, 2024, 8:02 p.m. UTC

On Sat, Mar 30, 2024 at 09:54:02AM +0100, Daniel Stenberg wrote:

> On Fri, 29 Mar 2024, Jeff King wrote:
> 
> > I noticed some http-related failures in the test suite on my Debian
> > unstable system, which recently got an upgraded curl package. It looks
> > like it's related to cases where we use the remote-curl helper in
> > "connect" mode (i.e., protocol v2) and the http buffer is small
> > (requiring us to stream the data to curl). Besides just running t5551,
> > an easy reproduction is:
> 
> This smells like a libcurl regression to me. I "imported" this into our
> issue tracker here: https://github.com/curl/curl/issues/13229

This was all resolved in that issue, but just to summarize for the list
here: it was a regression in curl and there's a fix already. Thanks
Daniel for your (as usual) prompt digging into the problem (and likewise
to Stefan for the actual fix).

Ultimately the issue will be fixed by moving to a different version of
libcurl, but here's an easy workaround in the meantime, with a small doc
cleanup I found along the way.

  [1/2]: http: reset POSTFIELDSIZE when clearing curl handle
  [2/2]: INSTALL: bump libcurl version to 7.21.3

 INSTALL | 2 +-
 http.c  | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

-Peff

Comments

Randall S. Becker April 2, 2024, 8:21 p.m. UTC | #1

On Tuesday, April 2, 2024 4:03 PM, Peff wrote:
>To: [e-mail]
>Cc: Daniel Stenberg <[e-mail]>
>Subject: [PATCH 0/2] git+curl 8.7.0 workaround
>
>On Sat, Mar 30, 2024 at 09:54:02AM +0100, Daniel Stenberg wrote:
>
>> On Fri, 29 Mar 2024, Jeff King wrote:
>>
>> > I noticed some http-related failures in the test suite on my Debian
>> > unstable system, which recently got an upgraded curl package. It
>> > looks like it's related to cases where we use the remote-curl helper
>> > in "connect" mode (i.e., protocol v2) and the http buffer is small
>> > (requiring us to stream the data to curl). Besides just running
>> > t5551, an easy reproduction is:
>>
>> This smells like a libcurl regression to me. I "imported" this into
>> our issue tracker here: https://github.com/curl/curl/issues/13229
>
>This was all resolved in that issue, but just to summarize for the list
>here: it was a regression in curl and there's a fix already. Thanks Daniel for your (as
>usual) prompt digging into the problem (and likewise to Stefan for the actual fix).
>
>Ultimately the issue will be fixed by moving to a different version of libcurl, but
>here's an easy workaround in the meantime, with a small doc cleanup I found along
>the way.
>
>  [1/2]: http: reset POSTFIELDSIZE when clearing curl handle
>  [2/2]: INSTALL: bump libcurl version to 7.21.3
>
> INSTALL | 2 +-
> http.c  | 1 +
> 2 files changed, 2 insertions(+), 1 deletion(-)

Do we have an ETA for this fix? That or do we know when curl is planning on resolving this?

Thanks,
Randall

Jeff King April 2, 2024, 8:31 p.m. UTC | #2

On Tue, Apr 02, 2024 at 04:21:56PM -0400, [e-mail] wrote:

> Do we have an ETA for this fix? That or do we know when curl is
> planning on resolving this?

The fix is in curl's master branch already. According to their
docs/RELEASE-PROCEDURE.md, the next release would be May 22. But Daniel
is the more authoritative word. ;)

-Peff