-
Type:
Bug
-
Status: Resolved
-
Priority:
Minor
-
Resolution: Fixed
-
Affects Version/s: None
-
Component/s: Tree Snapshot
-
Team:PLATFORM
-
Sprint:nxplatform #98
-
Story Points:1
The test already has a mechanism to retry before failing but it seems to be not enough as can be seen in 2021 and 2023 in the postgresql tests.
[2023-10-12T20:11:06.901Z] [ERROR] testSnapshotableVersionRemovalPolicy(org.nuxeo.snapshot.TestSnapshoting) Time elapsed: 2.042 s <<< FAILURE! [2023-10-12T20:11:06.901Z] java.lang.AssertionError: expected:<0> but was:<3> [2023-10-12T20:11:06.901Z] at org.junit.Assert.fail(Assert.java:89) [2023-10-12T20:11:06.901Z] at org.junit.Assert.failNotEquals(Assert.java:835) [2023-10-12T20:11:06.901Z] at org.junit.Assert.assertEquals(Assert.java:647) [2023-10-12T20:11:06.901Z] at org.junit.Assert.assertEquals(Assert.java:633) [2023-10-12T20:11:06.901Z] at org.nuxeo.snapshot.TestSnapshoting.testSnapshotableVersionRemovalPolicy(TestSnapshoting.java:356)
This random happens in the end of the test suite as the nuxeo-tree-snapshot-core module is run in the end. When happening in the retry as well, it's a 7h30 build.
The failing assertion could be looped a few times to prevent failing:
https://github.com/nuxeo/nuxeo-lts/blob/2021/modules/platform/nuxeo-tree-snapshot-core/src/test/java/org/nuxeo/snapshot/TestSnapshoting.java#L352-L356