Re: r-bioc-biocinstaller (Was: r-base breaks 9 autopkgtests)

[Graham Inggs <ginggs@debian.org>]

Hi Andreas

On 2019/07/26 14:59, Andreas Tille wrote:
> > OK, great!  As per the excuses for r-base [1], once the regressions in
> > r-cran-filehash and r-cran-haplo.stats have been fixed, I will ask
> > Dirk to do a no-change source-only upload to resolve the "not built on
> > buildd" issue.
>
> If those two packages are the only blockers we might consider
> deactivating the tests for the moment.  I'm to busy at DebConf to
> do more complex stuff than simple uploads / upgrades but feel free
> to "cheat" with the tests.  I just activated the tests which were
> never be run before so deactivating them is just restoring the
> situation we had before.  These should not block any transition.

I trust you had a good DebConf!

As far as I can see, these tests were activated and passed previously, 
which is why, as you can see on the packages tracker pages [1][2], 
britney considers them regressions, and thus block migration:

autopkgtest for r-cran-filehash/2.4-2-1: amd64: Regression
Not considered

autopkgtest for r-cran-haplo.stats/1.7.9-3: amd64: Regression
Not considered

I'll quote the usual message that appears on the package tracker for 
transitions:

"Please avoid uploads unrelated to this transition, they would likely 
delay it and require supplementary work from the release managers. On 
the other hand, if your package has problems preventing it to migrate to 
testing, please fix them as soon as possible."

The uploads we need now are ones fixing autopkgtest regressions and 
binaries not built on the buildds, as these are blocking the migration.

I found one can search on britney's update excuses page [3] e.g. for 
'binaries uploaded by ginggs' and find a package of mine that recently 
cleared NEW and was not built on the buildds.

Regards
Graham


[1] https://tracker.debian.org/pkg/r-cran-filehash
[2] https://tracker.debian.org/pkg/r-cran-haplo.stats
[3] https://release.debian.org/britney/update_excuses.html