lp:~paelzer/britney/+git/hints-ubuntu

Author: Christian Ehrhardt 
Author Date: 2022-11-29 06:55:50 UTC

badtest ubuntu-fan (LP: #1998184)

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2022-11-29 06:55:50 UTC

badtest ubuntu-fan (LP: #1998184)

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2022-11-29 06:55:50 UTC

badtest ubuntu-fan (LP: #1998184)

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2022-11-29 06:55:08 UTC

badtest ubuntu-fan (LP: #1998184)

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2022-11-29 06:53:09 UTC

badtest ubuntu-fan (LP: #1998184)

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2022-01-12 08:57:14 UTC

DPDK was removed on armhf

DPDK 21.11 intentionally removed armhf builds [1].
Remaining arches test fine e.g. [2] with 21.11 binaries, but on armhf
it now tests old binaries [3] which fails badly and even if working
would not test the version in proposed (as there is none for armhf).

Marked as force-badtest along the other similar cases in the
ubuntu-release hints file.

[1]: https://launchpad.net/ubuntu/+source/dpdk/21.11-1
[2]: https://autopkgtest.ubuntu.com/results/autopkgtest-jammy/jammy/arm64/d/dpdk/20220111_121315_99319@/log.gz
[3]: https://autopkgtest.ubuntu.com/results/autopkgtest-jammy/jammy/armhf/d/dpdk/20220111_122708_f3daf@/log.gz

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2021-12-13 10:00:39 UTC

reset postgresql-mysql-fdw (LP: #1954374)

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2021-11-11 10:07:45 UTC

bionic reset plv8 on ppc64 (LP: #1950268)

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2021-08-16 10:47:09 UTC

mark pglogical on armhf as badtest

This test is really bad anyway, but it turns out that on armhf after a
former reset it sometimes works and then again enters phases where it
times out without an obvious fixable reason.

Therefore badtest the current version on armhf, the others are still
correctlty handled by the former reset.

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2021-07-19 11:45:13 UTC

reset tests of xtensor on s390x/arm64/i386 (LP: #1936820)

Due to a change in the Debian packaging this isn't
working on s390x / arm64 / i386.
The xtensor-dev is now arch:all (as transitional) but depends on
libxtensor-dev which isn't available on varius architectures.

Instead of going into extensive fixes of the dependencies we can
as well reset the tests on those architectures. Once it builds
down the road we will automatically pick it up again for test coverage.

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2021-06-08 13:55:04 UTC

update i386 arch:all hint

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2021-06-07 05:18:02 UTC

reset puppetdb stuck in proposed and unstable (LP: #1885568)

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2021-05-31 14:05:14 UTC

postgresql: armhf test environment triggers permission issues

This is slightly odd as outlined in [1] it is only occurring in recent
bionic armhf tests. The root cause is clear and fixed later on via [2].

We don't want to push an SRU package update just for this self-test fixup
and in these packages case the tests on armhf cover exactly the same
as the other architectures which still work.

Therefor reset the affected tests on armhf in bionic

[1]: https://bugs.launchpad.net/ubuntu/+source/postgresql-12/+bug/1928773/comments/4
[2]: https://salsa.debian.org/postgresql/postgresql-common/-/commit/aa4829c899a3cf7ea6c90990d989dd57d2a63857

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2020-12-06 14:19:25 UTC

add open-iscsi to # i386 uninteresting as a host generally

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2020-12-07 14:59:36 UTC

postgresql: reset pgl-ddl-deploy that was broken in 2.x uploads

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2020-11-26 14:43:58 UTC

ubuntu-image broken by python-debian 0.1.38

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2020-11-26 13:40:22 UTC

omnidb 2.17.0+ds-2build1 can't work with postgresql-13

This is a bit of a circular dependency. We currently have:
 omnidb | 2.17.0+ds-2build1 | hirsute/universe | source
 omnidb | 3.0.2b+ds-1 | hirsute-proposed/universe | source

Of those 2.17.0+ds-2build1 can't work with the new combo of
postgresql-13/postgresql-common that I need to migrate.

But we can't "fix" omnidb either, since >=3.0 has dropped the test
entriely due o the fact that the plugin functionality was moved to a
different source.

Now we have a growing list of tests that is automatically retried vs the
version in -release [1].
But at the same time we can't make it "test against 3.x to realize
things are ok" as that will fail with "SKIP no tests in this package"
which is considered a "Neutral" state [2].

Given the current size of the ppc64 queue I'm proposing to ignore the
old test results - which will be more effective than triggering it to
report a "neutral" fail and then hoping this will make it pass.

[1]: https://autopkgtest.ubuntu.com/packages/o/omnidb/hirsute/ppc64el
[2]: https://objectstorage.prodstack4-5.canonical.com/v1/AUTH_77e2ada1e7a84929a74ba3b87153c0ac/autopkgtest-hirsute/hirsute/amd64/o/omnidb/20201124_070404_c8486@/log.gz

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2020-11-19 14:06:13 UTC

hint postgresql-multicorn until the transition to postgresql-13 is fully completed

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2020-11-18 13:58:00 UTC

reset pgpointcloud on armhf

pgpoincloud on armhf is known to have issues, see
  https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=974869

There is no fix yet, but the coverage of this exists on all other
architectures and resetting it would allows things to pass for now -
while automatically becoming a valid test once we get a fixed version.

We currently have a sync:
 pgpointcloud | 1.2.1-5 | hirsute-proposed/universe | source

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2020-11-10 15:09:11 UTC

reset shapeit4 (was a once in a lifetime pass)

The results of shapeit4 on s390x are continuously failing for years
https://autopkgtest.ubuntu.com/packages/s/shapeit4/hirsute/s390x
https://autopkgtest.ubuntu.com/packages/s/shapeit4/groovy/s390x
https://autopkgtest.ubuntu.com/packages/s/shapeit4/focal/s390x

We recently had an odd outlier that passed and therefore this now starts
to block transitions like boost 1.71

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2020-11-09 15:04:50 UTC

marisa: bump version

This was one of the i386 that has to be checked in more detail to
decide. That has not yet happened but since the version got a few buildX
suffixes the old override does not more apply.

Bump the version to the one currently in hirsute universe/proposed.

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2020-11-09 15:01:20 UTC

libftdi1: pkg-config fails on i386

The i386 tests fail as they need pkg-config to work.
libftdi1-dev:i386 installs fine and brings
/usr/lib/i386-linux-gnu/pkgconfig/libftdi1.pc but in our amd64-plus-i386
test environment that isn't considered and pkg-config fails.

Mark the test as not cross-testable for i386.

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>

Author: Christian Ehrhardt 
Author Date: 2020-11-04 09:58:47 UTC

bump apt-clone reset (LP: #1510060 LP: #1902830)

Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com>