On Wed, Apr 9, 2025 at 7:02 AM Santiago Vila <sanv...@debian.org> wrote: > Dear Release Managers: > > Based on this reply by Paul: > > https://lists.debian.org/debian-devel/2025/04/msg00065.html > > I'd like to request that the severity of the following bugs is raised > to serious again: > .. > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1057562 > > (Raised by Sebastian to serious, downgraded later by the maintainer) > > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1069402 > > This is essentially the same as #1057562, except that it's > for gcr version 3 which is in a different package. > > Note that both gcr and gcr4 are key packages, so a serious severity > will not make the package to be removed.
It feels like there is less reason to argue about severity for gcr & gcr4 then? > A brief explanation for the maintainers that we don't want > flaky tests in trixie would also be nice. "flaky" is vague and until there is a specific standard for failure rate, this is going to be subject to maintainer discretion to some degree. As of 4.4.0.1, gcr4 does not appear to be flaky at all in the official buildds or on the reproducible builds server: https://buildd.debian.org/status/package.php?p=gcr4 https://tests.reproducible-builds.org/debian/history/gcr4.html I cherry-picked the fix to gcr which now shows the same result. The particular test is not used in autopkgtests so that isn't a concern here. Thank you, Jeremy Bícha
