Hi Michael, On 22-05-2021 09:57, Michael Banck wrote: >> Your package has an autopkgtest, great. However, because of the failure >> "caused" by glbic, > > Can you elaborate what you mean by that, is that related to pymol? Or > are you saying that due to glbic (glibc?) failures, you had to look at > all flaky tests and pymol was one of them?
The last one (I looked at all failures, glibc typically lets me find flaky tests as they trigger so many tests), hence the caused between quotes. And indeed a typo: glibc. >> I looked into the history of your autopkgtest [1] and I noticed it >> fails regularly on arm64 because it hits the autopkgtest timeout after >> 2:47 hours. >> >> Because the unstable-to-testing migration software now blocks on >> regressions in testing, flaky tests, i.e. tests that flip between >> passing and failing without changes to the list of installed packages, >> are causing people unrelated to your package to spend time on these >> tests. > > Well I'm going to try to take a look, but this apparently being > arm64-specific won't make it easier. If I can help by extracting extra information from a failing run in one of arm64 workers, don't hesitate to tell me/us what you would need. Also, you could upload to experimental with extra debugging options (and potentially store stuff in the artifacts) and re-trigger the britney created runs for the pseudo excuses until you hit the failure mode. Paul
OpenPGP_signature
Description: OpenPGP digital signature