action #37354
closed
[opensuse][functional][u][sporadic][medium] test fails in desktop_runner is unstable
Added by JERiveraMoya almost 6 years ago.
Updated almost 6 years ago.
Category:
Bugs in existing tests
Target version:
SUSE QA - Milestone 17
Description
Observation¶
openQA test in scenario opensuse-Tumbleweed-DVD-x86_64-update_Leap_42.2_kde@64bit fails in
desktop_runner
After typing "true" does not come back to desktop.
Also fails when the box on the top does not dissapear and the needle in the desktop does not match: https://openqa.opensuse.org/tests/690612#step/desktop_runner/4
Reproducible¶
Fails since (at least) Build 20180608
Expected result¶
Last good: 20180606 (or more recent)
Acceptance criteria¶
- AC1: The test module "desktop_runner" should be made stable, e.g. less than 1/100 fails in "desktop_runner" -> AT1: Run 100 jobs to verify, could be on o3
Further details¶
Always latest result in this scenario: latest
- Related to coordination #35302: [qe-core][opensuse][functional][epic][sporadic] Various unstable tests on o3 added
- Subject changed from [opensuse][functional][sporadic] test fails in desktop_runner is unstable to [opensuse][functional][u][sporadic] test fails in desktop_runner is unstable
- Description updated (diff)
- Due date set to 2018-07-03
- Status changed from New to Workable
- Target version set to Milestone 17
- Target version changed from Milestone 17 to Milestone 17
- Subject changed from [opensuse][functional][u][sporadic] test fails in desktop_runner is unstable to [opensuse][functional][u][sporadic][medium] test fails in desktop_runner is unstable
- Status changed from Workable to In Progress
- Assignee set to zluo
after investigation and run more than 40 test runs on normal remoter workers and remote workers (loewe), I can see that this issue is related to low performance on server.
So need to workaround issue with send_key 'ret' in sub x11_start_program of susedistribution.pm.
PR:
https://github.com/os-autoinst/os-autoinst-distri-opensuse/pull/5275
I don't know a better way to workaround issue related on openQA production server. The statistic shows only problem with low performance on server.
I am sorry but repeatedly hitting the same key does not sound like a good solution. If the problem can only be reproduced in production for now then I recommend to gather good statistics on production. https://progress.opensuse.org/projects/openqatests/wiki/Wiki#Statistical-investigation mentions one approach for that. We had pretty stable tests for a while so either the product degraded in stability or we have unstable openQA workers. Both needs to be fixed not in os-autoinst-distri-opensuse. I do not consider it a long-term scalable approach to provide quirky workarounds in tests.
- Status changed from In Progress to Workable
- Status changed from Workable to In Progress
- Assignee set to SLindoMansilla
The last round of statistical investigation shows that problem happens 45 times of 100 jobs.
Ok, with this analysis as written in #35685#note-32 I guess it is better we work on #31351 first. It would be best you pair up with zluo on this.
- Blocked by action #31351: [functional][u][medium] force_cron_run does not actually run any crons (occasionally) added
- Status changed from In Progress to Blocked
- Assignee deleted (
SLindoMansilla)
@SLindoMansilla please don't keep unassigned blocked tickets. It's ok if you want to assign to me.
- Due date changed from 2018-07-03 to 2018-07-17
Same as #35685, let's see about stability after #31351 was resolved now.
- Description updated (diff)
- Status changed from Blocked to Resolved
Seems we covered this now properly with #35685 -> desktop_runner stable
- Related to action #39011: [functional] Low performance on openqa production server added
Also available in: Atom
PDF