| Summary: | test_rpm_query_nonroot fails to delete user | ||||||
|---|---|---|---|---|---|---|---|
| Product: | [QA/Testing] Runtime Testing | Reporter: | Alejandro <alejandro> | ||||
| Component: | oeqa/runtime | Assignee: | Pavel Zhukov <pavel> | ||||
| Status: | RESOLVED FIXED | QA Contact: | |||||
| Severity: | normal | ||||||
| Priority: | Medium | CC: | luca.ceresoli, randy.macleod, steve | ||||
| Version: | unspecified | ||||||
| Target Milestone: | 4.2 | ||||||
| Hardware: | x86 | ||||||
| OS: | Multiple | ||||||
| Whiteboard: | AB-INT | ||||||
| OS type for building Yocto: | --- | Type of Regression: | --- | ||||
| Verified: | Documentation change: | No (bug/feature does not impact docs) | |||||
| Attachments: |
|
||||||
|
Description
Alejandro
2021-04-14 22:57:00 UTC
Happens on dunfell and qemuppc Another occurrence on dunfell (ubuntu1804-ty-1, qemuppc-alt) https://autobuilder.yoctoproject.org/typhoon/#/builders/107/builds/1625 We've done 100s of builds since this was last seen in May so closing. Happened again today :-( https://autobuilder.yoctoproject.org/typhoon/#/builders/107/builds/3621/steps/12/logs/stdio qemuppc-alt on debian11-ty-3 Created attachment 4906 [details]
Full do_testimage log
Probably a race condition in rpm.RpmBasicTest.test_rpm_query_nonroot. unset_up_test_user() waits up to 30 seconds to "ensure no test1 process in running", then it fails with this error. Also on master. backports dunfell and kirkstone Pavel, are you able to work on this bug or should I move it to unassigned? (In reply to comment #9) > Pavel, are you able to work on this bug or should I move it to unassigned? Hi Randy, I'm able to work on this bug. Need few days to sort few things out Looks like we hit this: https://github.com/systemd/systemd/issues/8598 There are few options atm: - Increase timeout (30 secs currently) - # systemctl mask user@.service - I'm not sure that will be consequences of disabling user services on project level - force removal of the user (with "-f" option) - Wait for systemd bug to be closed (or fix it :) ) Hopefully "fixed". Feel free to reopen if happens again |