CI: Sometimes "TestSynchronized" test cases fail in the CI Job
Parent: [EPIC] Playout Stability (aura#406)
When repeating the job they usually succeed.
Example where it failed on 1st try: https://gitlab.servus.at/aura/engine/-/jobs/6945
Un-comment test case test_synchronized
.
Designs
- Show closed items
Related branches 1
Activity
-
Newest first Oldest first
-
Show all activity Show comments only Show history only
- David Trattnig changed milestone to %1.0-alpha2 —
Precise Pheasantchanged milestone to %1.0-alpha2 —
Precise Pheasant - David Trattnig added P2 bug labels
- David Trattnig assigned to @david
assigned to @david
- David Trattnig changed the description
changed the description
- David Trattnig mentioned in commit f879048f
mentioned in commit f879048f
- David Trattnig changed milestone to %1.0-alpha5 — Capable Capybara
changed milestone to %1.0-alpha5 — Capable Capybara
- Author Owner
Also this one fails from time to time
TestEngineExecutor.test_parent_child_replacement_in_time
.Edited by David Trattnig - David Trattnig added #127 (closed) as child task
added #127 (closed) as child task
- David Trattnig mentioned in commit 856ce196
mentioned in commit 856ce196
- David Trattnig changed milestone to %1.0-alpha3 — Playful Platypus
changed milestone to %1.0-alpha3 — Playful Platypus
- David Trattnig assigned to @chrizz and unassigned @david
- David Trattnig marked this issue as related to #137 (closed)
marked this issue as related to #137 (closed)
- David Trattnig added 1 deleted label
added 1 deleted label
- David Trattnig added To Do label
added To Do label
- David Trattnig added 1 deleted label
added 1 deleted label
- Chris Pastl added Doing label and removed To Do label
- Maintainer
As discussed, I enabled
test_synchronized
and added a loop to increase probability of a failure.I'm not sure if it's sufficient to do the loop inside the py files, so I modified the makefile in order to execute the target
coverage
multiple (e.g. 100) times... - Author Owner
Any news on this, can it be closed?
- Maintainer
I tried different loop syntaxes which should work in makefiles and also worked on my local machine but for some reasons, the testcases were executed only once in the ci.
Anyhow, the tests ran very often in the past months and I also tried to trigger the error manually - I didn't notice any failure. Thus closing this ticket.
Edited by Chris Pastl