Merge lp:~paelzer/britney/hints-ubuntu-disco-fix-systemd-ppc-hint-that-never-works into lp:~ubuntu-sru/britney/hints-ubuntu-disco

Proposed by Christian Ehrhardt 
Status: Merged
Merged at revision: 3711
Proposed branch: lp:~paelzer/britney/hints-ubuntu-disco-fix-systemd-ppc-hint-that-never-works
Merge into: lp:~ubuntu-sru/britney/hints-ubuntu-disco
Diff against target: 19 lines (+6/-3)
1 file modified
ubuntu-sru (+6/-3)
To merge this branch: bzr merge lp:~paelzer/britney/hints-ubuntu-disco-fix-systemd-ppc-hint-that-never-works
Reviewer Review Type Date Requested Status
Łukasz Zemczak not all of the changes Approve
Review via email: mp+373200@code.launchpad.net
To post a comment you must log in.
3712. By Christian Ehrhardt 

mark systemd broken on armhf

In the discussions in LP: #1845337 we realized that since recent
changes [1] to the infrastructure new directories are exposed [2].
Due to that this test is broken on LXD which currently means on armhf.

A following systemd upload could fix the test by adding the
directories to the list of filtered paths, but for mask the
bug as known bad test.

[1]: https://git.launchpad.net/autopkgtest-cloud/commit/?id=b8c9165686c7598b3f1a68aa4684e7f382ad935c
[2]: https://bugs.launchpad.net/ubuntu/+source/qemu/+bug/1845337/comments/10

Revision history for this message
Christian Ehrhardt  (paelzer) wrote :

Extended the MP by the change needed for [1] until resolved in systemd or reverted in the autopkgtest infrastructure.

[1]: https://bugs.launchpad.net/ubuntu/+source/qemu/+bug/1845337

Revision history for this message
Łukasz Zemczak (sil2100) wrote :

Ok, so generally I would be -1 on setting this to a generalized hint. Adding a generic hint usually means "this is broken and we do not care about it ever again", which means we basically have 0 testing for the given arch. The need to bump the test every time, even though troublesome, does require someone to take a look and confirm if the same thing is broken as it was before. Sure, this is very annoying, but I wouldn't want us to completely stop caring about testing for a given arch.

That being said, do not treat my words as the definite 'no'. I completely understand how unproductive this currently is, and I myself am putting my own rationale in doubt, having mixed feelings if it actually matters that much. If there are other SRU members that would be +1 on making this generic for disco, I would not object.

So for now I will merge your change, but not commit the actual generalization. I'll try discussing this with other SRU members on our nearest meeting.

review: Approve (not all of the changes)
Revision history for this message
Christian Ehrhardt  (paelzer) wrote :

Thanks!
I'm usually also rather safe and formal and wouldn't have requested that without a reason.
But to me [1][2] really seems like people have "having given up" on it.

Just like you I'm open to a compromise as well, I'm fine if we add this only versioned but I wanted to have exactly that SRU Team discussion that you mentioned.

[1]: http://autopkgtest.ubuntu.com/packages/s/systemd/disco/ppc64el
[2]: http://autopkgtest.ubuntu.com/packages/s/systemd/disco/armhf

Preview Diff

[H/L] Next/Prev Comment, [J/K] Next/Prev File, [N/P] Next/Prev Hunk
1=== modified file 'ubuntu-sru'
2--- ubuntu-sru 2019-09-19 15:11:55 +0000
3+++ ubuntu-sru 2019-09-26 14:51:40 +0000
4@@ -16,9 +16,12 @@
5 # Chrony regressed in release - fixes are on the way (LP: #1836929)
6 force-badtest chrony/3.4-1ubuntu1
7
8-# Regressed in release
9-force-badtest systemd/240-6ubuntu5.3/ppc64el
10-force-badtest systemd/240-6ubuntu5.4/ppc64el
11+# Regressed in release and does not come back
12+force-badtest systemd/all/ppc64el
13+# Regressed due to LXD nesting exposing unexpected directories (LP: #1845337)
14+# Triggering now due to:
15+# https://git.launchpad.net/autopkgtest-cloud/commit/?id=b8c9165686c7598b3f1a68aa4684e7f382ad935c
16+force-badtest systemd/240-6ubuntu5.7/armhf
17
18 # blacklisted because the autopkgtest installs not freely usable software
19 force-badtest virtualbox-ext-pack/all

Subscribers

People subscribed via source and target branches