On 7/13/21 1:10 PM, Ed Santiago wrote:
It's a flake: https://github.com/containers/crun/issues/660
That flake was claimed to have been closed in crun back in May.
Does this CI system have a crun later than May 10?
Oh good, I'm glad I asked. Ugh, that looks like a race condition that
really shouldn't be exposed to mortals.
That system (Note: F33) had crun-0.19.1-3.fc33-x86_64 at the time:
Conveniently I have that package installed on my laptop, an rpm -qi shows:
Build Date : Mon 17 May 2021 10:18:30 AM EDT
Ooof! I don't fully understand where the race is from the issue, and it
doesn't seem clear (to my ignorant eyes) how the fix works. Could it be
a related/unconsidered failure-path maybe? Should we re-open the issue
and rope in the experts?
Chris Evich (he/him), RHCA III
Senior Quality Assurance Engineer
If it ain't broke, yain't tryin' hard 'nough.