On Thu, Nov 06, 2025 at 11:02:47PM +0100, Simon Josefsson wrote: > forwarded 1115140 https://codeberg.org/shepherd/shepherd/issues/20 > thanks > > Thanks for report and sorry for these repeated failures here. I made an > upload to ignore failures for three self-tests which we have repeated > known and forwarded failures for, but still running them during build to > retain failure logs for upstream to look at. Alas I don't have time > right now to debug these further.
Hi. Thanks for handling this so quickly! I've visited the above URL and there is a comment saying this happens on amd64 with single CPU, but nowhere I said that was the case. (You can look at the build logs I provided, they have both DEB_BUILD_OPTIONS=parallel=2). I tried with machines with 1 and 2 CPUs, and my experience is that the build fails when using sbuild + schroot with the file backend, regardless of number of CPUs, and the build succeeds when using sbuild + unshare backend, also regardless of number of CPUs. (So, if you could clarify this in the upstream bug tracker, I think it would be good to not mislead the authors). Note that I fully support what you did, namely, disabling the test and let upstream to care about debugging. Thanks.
