Monday, 15 February 2016

Re: ANN: autopkgtests are now specific for "triggering" package for better isolation

On 10.11.2015 14:55, Martin Pitt wrote:
> Hello all,
> TL;DR: This is mostly of interest for people looking at/maintaining
> autopkgtests and update-excuses.html or retrying failed tests.
> With our current approach to running autopkgtests it sometimes happens
> that a package P in -proposed gets stuck because one of its
> dependencies D also gets uploaded and is broken.
> Right now we always run all tests against the entirety of -proposed,
> which will cause P's tests to fail due to the broken D-proposed.
> However, we can do better than that: When evaluating P, we should run
> P-proposed packages against D-release if possible, so that a broken
> D-proposed can't cause P-proposed to be held back. Conversely, for
> evaluating D-proposed all tests should be run with P-release and the
> binaries of D-proposed.

I'm unhappy about that. This is a major drawback trying to get depending
transitions migrated.

Currently looking at getting netcdf, openmpi, libjsoncpp, python-numpy,
scalapack, petsc, slepc, vtk6 and a dozen others migrated.

- so you need to check update_excuses *manually* for failing
tests, figure out that the test failure is caused by a
dependency on an outdated package in the release pocket,
report it, ask for a give back ...

- you have to repeat that again, if a test is triggered again.

Trying to get the above packages transitioned for two weeks now, this is kind of

Britney already knows about the depends, e.g.

vtk6 (6.2.0+dfsg1-4ubuntu6 to 6.2.0+dfsg1-8)
Depends: vtk6 libjsoncpp (not considered)
Depends: vtk6 netcdf
Depends: vtk6 openmpi (not considered)

so the autopkg tests should not even try to test with these packages from the
release pocket.

Martin suggested on irc to fix the dependencies of the packages instead to
(build-)depend on the version in proposed, however this would add some thousand
deltas to the packaging, which isn't sustainable.


ubuntu-devel mailing list
Modify settings or unsubscribe at: