Bug#1108435: unblock: nftables/1.1.2-1
Control: tags -1 moreinfo confirmed
On 2025-06-29 10:09:00 +0100, Jeremy Sowden wrote:
> On 2025-06-28, at 17:31:19 +0200, Sebastian Ramacher wrote:
> > On 2025-06-28 15:10:41 +0100, Jeremy Sowden wrote:
> > > Package: release.debian.org
> > > Severity: normal
> > > X-Debbugs-Cc: nftables@packages.debian.org, azazel@debian.org
> > > Control: affects -1 + src:nftables
> > > User: release.debian.org@packages.debian.org
> > > Usertags: unblock
> > >
> > > Please unblock package nftables
> > >
> > > [ Reason ]
> > > The release of nftables currently in testing, 1.1.2, introduces a
> > > regression. Upstream put out a new release, 1.1.3, shortly afterwards to
> > > fix this bug and two others. We would like to get these fixes into Trixie.
> > >
> > > See also https://bugs.debian.org/1107417
> > >
> > > [ Impact ]
> > > Two existing bugs will remain unfixed and a new one will be introduced.
> > >
> > > [ Tests ]
> > > Two of the three fixes are covered by the upstream test-suites.
> > >
> > > [ Risks ]
> > > The only change to the packaging is the update to d/changelog. The 1.1.3
> > > upstream release itself only contains these three bug-fixes:
> > >
> > > $ git log --oneline v1.1.2..v1.1.3
> > > 4431c70ca96c (tag: v1.1.3) build: Bump version to 1.1.3
> > > 2fbade3cd990 netlink: bogus concatenated set ranges with netlink message overrun
> > > 2c41c82296c7 evalute: make vlan pcp updates work
> > > 63ed8af82ba0 Revert "intervals: do not merge intervals with different timeout"
> > >
> > > so the risks should be minimal.
> > >
> > > [ Checklist ]
> > > [x] all changes are documented in the d/changelog
> > > [x] I reviewed all changes and I approve them
> > > [x] attach debdiff against the package in testing
> >
> > The debdiff seems to be missing.
>
> Yeah, I spotted that and sent it later. It's in the bug report:
>
> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1108435#17
>
> but for some reason it didn't make it to the mailing-list.
That is a good sign that the upload is not aligned with freeze policy.
But in this case, it is actually because of this:
> > > [ Other info ]
> > > When the upstream source tar-ball for 1.1.2 was prepared, a number of
> > > output files generated by running the test-suites were inadvertently
> > > included. The 1.1.3 tar-ball does not contains this cruft. The upshot is
> > > that while the source-code changes are limited, there is a lot of noise in
> > > the debdiff because of the removal of the junk files.
In such a case, feel free to filter the debdiff to make it reviewable.
In any case, please go ahead. When the upload is available in unstable,
please remove the moreinfo tag.
Cheers
--
Sebastian Ramacher
Reply to: