Bug#921566: Reopen Bug#921566 closed by Andreas Tille (Bug#921566: fixed in deepnano 0.0+git20170813.e8a621e-3)
Hi Paul, I recently had a look at deepnano and bug #921566, about the test crashing in timeout. I modified the test script to cap the total number of OpenMP threads, in hope this helps in case this is an improper handling of very large cores count, and I also enforced a ten minutes timeout on each command. I'm hopeful the deepnano package will not hog the CI infrastructure with these safeguards. Would it be possible to lift the ban on the package, once the version 0.0+git20170813.e8a621e-4 hits the archive? Thanks for caring for the CI infrastructure! Have a nice day, :) -- Étienne Mollier Fingerprint: 8f91 b227 c7d6 f2b1 948c 8236 793c f67e 8f0d 11da Sent from /dev/pts/5, please excuse my verbosity. On air: Threshold - Defence Condition signature.asc Description: PGP signature
Bug#921566: Reopen Bug#921566 closed by Andreas Tille (Bug#921566: fixed in deepnano 0.0+git20170813.e8a621e-3)
Hi Andreas, On 13-12-2019 17:24, Andreas Tille wrote: > However, even > > https://ci.debian.net/packages/p/deepnano > > does not exist any more. Am I missing something? A typo: https://ci.debian.net/packages/d/deepnano/ However, because of the shear amount of tests and size of the logs, we have to delete logs older than 3 months. There isn't anything to find here anymore. Paul signature.asc Description: OpenPGP digital signature
Bug#921566: Reopen Bug#921566 closed by Andreas Tille (Bug#921566: fixed in deepnano 0.0+git20170813.e8a621e-3)
Hi Paul, in our advent bug squashing party I wanted to have another look into this. On Sat, Feb 09, 2019 at 01:23:22PM +0100, Paul Gevers wrote: > > I admit I see no better solution than to deactivate the autopkgtest. > > That would really be a shame. There is really a regression in you > package, as the test used to finish in within two minutes and now times > out after 2:47 hours. So something broke it, albeit maybe not in your > setup. If you remove the autopkgtest to "fix" this bug, I suggest to not > close this bug, but retitle it to something like "deepnano is not > finishing properly and/or in time in some environments" and leave it > open until the issue is resolved (by deepnano, or by the (indirect) > dependency that caused this issue). It's clear caused by something > outside of deepnano, maybe > https://ci.debian.net/data/packages/unstable/amd64/d/deepnano/1680359.log has > a hint for you? However, even https://ci.debian.net/packages/p/deepnano does not exist any more. Am I missing something? Kind regards Andreas. -- http://fam-tille.de
Bug#921566: Reopen Bug#921566 closed by Andreas Tille (Bug#921566: fixed in deepnano 0.0+git20170813.e8a621e-3)
Hi Andreas, On 08-02-2019 21:17, Andreas Tille wrote: > I admit I see no better solution than to deactivate the autopkgtest. That would really be a shame. There is really a regression in you package, as the test used to finish in within two minutes and now times out after 2:47 hours. So something broke it, albeit maybe not in your setup. If you remove the autopkgtest to "fix" this bug, I suggest to not close this bug, but retitle it to something like "deepnano is not finishing properly and/or in time in some environments" and leave it open until the issue is resolved (by deepnano, or by the (indirect) dependency that caused this issue). It's clear caused by something outside of deepnano, maybe https://ci.debian.net/data/packages/unstable/amd64/d/deepnano/1680359.log has a hint for you? Paul signature.asc Description: OpenPGP digital signature
Bug#921566: Reopen Bug#921566 closed by Andreas Tille (Bug#921566: fixed in deepnano 0.0+git20170813.e8a621e-3)
Hi Paul, On Fri, Feb 08, 2019 at 09:02:23PM +0100, Paul Gevers wrote: > Control: reopen -1 > > On 07-02-2019 09:45, Debian Bug Tracking System wrote: > > deepnano (0.0+git20170813.e8a621e-3) unstable; urgency=medium > > . > >* Use less data for autopkgtest > > Closes: #921566 > > This seems to be not enough to fix this issue. I didn't lift the ban, > but trigger one run on our infrastructure: > > https://ci.debian.net/data/autopkgtest/unstable/amd64/d/deepnano/1874130/log.gz I admit I see no better solution than to deactivate the autopkgtest. Kind regards Andreas. -- http://fam-tille.de
Bug#921566: Reopen Bug#921566 closed by Andreas Tille (Bug#921566: fixed in deepnano 0.0+git20170813.e8a621e-3)
Control: reopen -1 On 07-02-2019 09:45, Debian Bug Tracking System wrote: > deepnano (0.0+git20170813.e8a621e-3) unstable; urgency=medium > . >* Use less data for autopkgtest > Closes: #921566 This seems to be not enough to fix this issue. I didn't lift the ban, but trigger one run on our infrastructure: https://ci.debian.net/data/autopkgtest/unstable/amd64/d/deepnano/1874130/log.gz Paul signature.asc Description: OpenPGP digital signature
