__main__.SeccompTest is failing on Ubuntu CI

Bug #1831296 reported by Evgeny Vereshchagin
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
systemd (Ubuntu)
Fix Released
Low
Unassigned
Eoan
Fix Released
Low
Unassigned

Bug Description

Since https://github.com/systemd/systemd/pull/12430 was merged and libsecomp was updated the test has been failing on Ubuntu CI: https://github.com/systemd/systemd/issues/12709. By analogy with https://github.com/systemd/systemd/pull/12430/commits/c3ab2c389ee60d92fb8d7fe779ae9c4e3c092e4c, the test should look for either "killed" or "dumped".

Revision history for this message
Dan Streetman (ddstreet) wrote :

Since https://github.com/systemd/systemd/pull/12430 isn't in any current Ubuntu systemd, I don't believe the test case needs to be fixed in any releases besides Eoan (so the test will already be corrected once Ubuntu systemd is updated to include pull 12430). Right?

And it should be corrected in:
https://salsa.debian.org/systemd-team/systemd/blob/master/debian/tests/boot-and-services

Right?

no longer affects: systemd (Ubuntu Bionic)
Revision history for this message
Evgeny Vereshchagin (evvers) wrote :

That's correct. The assertion is at https://salsa.debian.org/systemd-team/systemd/blob/master/debian/tests/boot-and-services#L421.

Regarding which release of Ubuntu is affected, I have to say I don't know. The issue is mostly about "upstream" mode of Ubuntu CI, which can be detected by checking an environment variable called TEST_UPSTREAM, as far as I can remember.

Revision history for this message
Evgeny Vereshchagin (evvers) wrote :

I'd add that it seems to me the "upstream" mode of Ubuntu CI doesn't seem to be actively maintained because unlike the other CI systems used upstream (which are usually fixed almost immediately), it can be broken for weeks (or sometimes even months) which isn't suitable for CI. In principle, if nobody takes care of it, it would probably makes sense to turn off Ubuntu CI upstream (which I did) and switch to the usual release cycle where maintainers try to release the package, catch a lot of bugs (easily preventable by the CI when PRs are opened) and report them upstream. It's far from ideal, of course, but I don't know what else to do about it.

Revision history for this message
Dan Streetman (ddstreet) wrote :
Changed in systemd (Ubuntu Eoan):
assignee: nobody → Dan Streetman (ddstreet)
importance: Undecided → Low
status: New → In Progress
Dan Streetman (ddstreet)
tags: added: ddstreet-next systemd
Revision history for this message
Martin Pitt (pitti) wrote :

Thanks Dan! I landed your PR, so it should apply to the next upstream CI run.

Changed in systemd (Ubuntu Eoan):
status: In Progress → Fix Committed
Dan Streetman (ddstreet)
tags: removed: ddstreet-next systemd
Changed in systemd (Ubuntu Eoan):
assignee: Dan Streetman (ddstreet) → nobody
Revision history for this message
Dan Streetman (ddstreet) wrote :
Changed in systemd (Ubuntu Eoan):
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.