On 2019-09-04 21:08, Paul Gevers wrote: > Hi Aurelien, > > On 02-09-2019 21:43, Aurelien Jarno wrote: > > [...] > > >>> - nemo: seems to be a bug in udisks2 postinst script > >>> - pg-repack: debci issue, the regression is caused by libpq5 from > >>> experimental, which is wrongly selected instead of the unstable > >>> version. > >>> - repmgr: ditto with libpq-dev. > > [...] > > >>> There are also a few failures that seems to be related to slight change > >>> of precision of the math functions: pyfai, sextractor. > >> > >> So what do you propose with these? Do they need to fix something? E.g. > >> their autopkgtest? Then we better warn them *now*, i.e. filing bugs. > > > > For those I still only have look the test logs. I still need to debug > > the issue first and also check if this is a real bug or a false > > positive. > > pyfai dropped of the list. > > Myon is aware of the postgresql-hll regression: > https://github.com/citusdata/postgresql-hll/issues/67 > > We'll remove the package from testing when we start the transition.
Thanks for the info. > I'll check why there are so many "Test in progress" items. Most of them > should be neutral I see. Ok. I have checked the one marked as regression, and there are only two if them that are real regressions in addition to postgresql-hll: - iraf-fitsutil: this is a strange issue, I am able to reproduce it, but I haven't investigated more yet. - sextractor: I am able to reproduce the issue, it seems to be related to a math precision change, but I haven't investigated more yet. Aurelien -- Aurelien Jarno GPG: 4096R/1DDD8C9B aurel...@aurel32.net http://www.aurel32.net
signature.asc
Description: PGP signature
