1:0.4.7 FTBFS in wily-proposed

I can build the same code in a vivid chroot without seeing any failures. My guess is that something else (python-apt? python-aptdaemon?) changed in the archive.

On Aug 21, 2015, at 04:07 PM, Alberto Milone wrote:

>I can build the same code in a vivid chroot without seeing any failures.
>My guess is that something else (python-apt? python-aptdaemon?) changed
>in the archive.

Oh yes, they surely have.

 python-apt | 0.9.3.11build1 | vivid | source, amd64, arm64, armhf, i386, powerpc, ppc64el
 python-apt | 1.0.0~beta3.1ubuntu1 | wily | source, amd64, arm64, armhf, i386, powerpc, ppc64el

 python-aptdaemon | 1.1.1+bzr982-0ubuntu3 | vivid | all
 python-aptdaemon | 1.1.1+bzr982-0ubuntu3.1 | vivid-security | all
 python-aptdaemon | 1.1.1+bzr982-0ubuntu3.1 | vivid-updates | all
 python-aptdaemon | 1.1.1+bzr982-0ubuntu8 | wily | all

I pushed https://github.com/tseliot/ubuntu-drivers-common/commit/0be1e7450a6a6d which resolves most of the failures. One failure left now: ubuntu_drivers.DetectTest.test_system_driver_packages_detect_plugins

Yep. 1:0.4.8 is still ftbfs.

https://github.com/tseliot/ubuntu-drivers-common/commit/7fd400283d fixes the long-standing race condition in the tests. That has been around for a while, causing random FTBFS, but it became much more frequent with current Python.

This bug was fixed in the package ubuntu-drivers-common - 1:0.4.10

---------------
ubuntu-drivers-common (1:0.4.10) wily; urgency=medium

  * debian/tests/system: Drop the nvidia package tests; they are now tested
    through autodep8 on every kernel and dkms change, so we don't have to rely
    on this hard to maintain autopkgtest any more. Also drop most other driver
    package tests for the same reason; just keep the "fglrx" test to get a
    full end-to-end test with "ubuntu-drivers".

 -- Martin Pitt <email address hidden> Thu, 03 Sep 2015 11:27:33 +0200