On Fri, Feb 10 2023, Junio C Hamano wrote: > Junio C Hamano <gitster@xxxxxxxxx> writes: > >> Ævar Arnfjörð Bjarmason <avarab@xxxxxxxxx> writes: >> >>>> I wonder if something like this is in order? >>> >>> I don't have much to contribute on that front, but this is really >>> missing some "why", this worked before, why is it failing now? Do we >>> have any idea. >> >> Your guess is as good as mine. I do not do Windows. > > This morning, I notice that those CI jobs I ran last night that > failed with "whoa, windows tests are somehow reporting that symlinks > are available but not really" issue the patch in this thread were > attempting to address has passed even for branches like 'master' and > 'next' that do not yet have it, and it seems to be because you > re-run these failed jobs. > > Whatever magic you used to fix these failing tests, thanks. > > Do you have an insight on why and how these were failing? The patch > in this thread was a band-aid without knowing why all of a sudden > "ln -s x y && test -h y" started passing (while compat/mingw.c still > says readlink() is not supported). If we know that such a breakage > is not expected, we can drop this workaroun, which would be great. I'm not Johannes :) But as I noted upthread this failed when we went from: Download action repository 'git-for-windows/setup-git-for-windows-sdk@v1' (SHA:cbe017cd7ae39629bf4e34fce8b1ccd211fec009) To: Download action repository 'git-for-windows/setup-git-for-windows-sdk@v1' (SHA:848609620edfa4c2fc64838b85fbe19e534236ee) And now our passing "next" has: Download action repository 'git-for-windows/setup-git-for-windows-sdk@v1' (SHA:a8e2a23eb07129d628ff6f9d5f11047b0662aeba) If you then look at that range in that repository you'll find the release includes: https://github.com/git-for-windows/setup-git-for-windows-sdk/pull/595 https://github.com/git-for-windows/setup-git-for-windows-sdk/pull/596 https://github.com/git-for-windows/setup-git-for-windows-sdk/pull/597 The release notes for the last of those then has: "Fix issue with symlink restoration on windows". Which from some looking around seems like it might be the issue we've been seeing, it's this commit & PR: https://github.com/actions/toolkit/commit/b2d865f18051b214475dae41766e9970fd68ca12 https://github.com/actions/toolkit/pull/1291 And following that rabbit hole leads to Johannes noting that this (or a related change) was breaking GFW: https://github.com/actions/toolkit/pull/1291/commits/2867e318d4d0de11b10a2887fb29dcf713559a71#r1098571737 So I think we can drop the workaround, at least as far as fixing the CI breakages goes.
