-
Bug
-
Resolution: Won't Do
-
Minor
-
None
-
rhel-9.2.0
-
None
-
None
-
rhel-net-mgmt
-
ssg_networking
-
None
-
False
-
False
-
-
No
-
None
-
None
-
None
-
If docs needed, set a value
-
-
Unspecified
-
None
-
0
The test is here: https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/blob/9934be52cd453202248cb5f7f7dadbe05a2412de/src/core/tests/test-l3cfg.c#L340
The test exists for a while, and it appeared to run stable. No issues in gitlab-ci, in copr builds, in beaker builds (or on my machine).
Note that this is a randomize test, that means it takes different code paths depending on NMTST_SEED_RAND environment variable. Also note that on my machine I always run unit tests with randomization enabled (by setting the seet to empty `NMTST_SEED_RAND=`). I didn't notice a failure so far.
However, in the gitlab-ci test script there is a bug, so that the test effectively run always with the same seed `NMTST_SEED_RAND=0`. This is about to be fixed, and the test start failing with certain randomizations.
This needs fixing.
Here is a run that shows the failure: https://gitlab.freedesktop.org/NetworkManager/NetworkManager/-/pipelines/794971
In the meantime, I disabled the test. To reproduce, you first need to re-enable it. If you cannot reproduce on your machine, try in gitlab-ci.
- external trackers