So it looks like the regression started about two weeks ago, in oa3e6c0 of the /etc/.git repo. I'm now looking into a few other suspect things which got subtracted during that commit as well.   Dave  ---- On Thu, 13 Nov 2025 09:00:49 -0500 David Cooper <david.cooper@genworks.com> wrote --- To be clear "we" = Georgiy, in the case of last night's quick triage, diagnosis, and fix.  What Georgiy didn't bother to mention was that the "issue" was a regression caused by my prompting an AI to regenerate a whole config file, rather than asking it for patches and applying them manually as I normally do, and as I was duly lectured to do by Georgiy himself at ELS in Zürich (well, copy/paste over entire files is what he explicitly lectured me _not_ to do, and that exact misbehavior on my part is what ended us up in this code 500 snafu). Hopefully I have learned my lesson. Thanks Georgiy, Dave ---- On Thu, 13 Nov 2025 04:51:46 -0500 Georgiy Tugai < mailto:georgiy@tugai.id.au > wrote --- Hi Robert, We fixed this issue last night; your CI pipeline still fails but it's not due to 500 errors anymore. Regards, Georgiy On 2025-11-10 21:11, Robert Goldman wrote: I was looking at a merge request for ASDF, and it failed all the CI/CD tests. See, for example, https://gitlab.common-lisp.net/rbrown/asdf/-/jobs/103352 What's odd is that it looks like this failure isn't about anything failing, it's about CI/CD not being able to clone the git repo. Getting source from Git repository 00:00 Gitaly correlation ID: 01K9DE5HXE6N8KDGC4YQ4ERFSV Fetching changes... Initialized empty Git repository in /builds/rbrown/asdf/.git/ Created fresh repository. fatal: unable to access ' https://gitlab.common-lisp.net/rbrown/asdf.git/ ': The requested URL returned error: 500 Any idea what is going haywire here? Thanks! R