Closed
Labels
priority: p2Moderately-important priority. Fix may not be included in next release.Moderately-important priority. Fix may not be included in next release.type: bugError or flaw in code with unintended results or allowing sub-optimal usage patterns.Error or flaw in code with unintended results or allowing sub-optimal usage patterns.
Metadata
Metadata
Assignees
Labels
priority: p2Moderately-important priority. Fix may not be included in next release.Moderately-important priority. Fix may not be included in next release.type: bugError or flaw in code with unintended results or allowing sub-optimal usage patterns.Error or flaw in code with unintended results or allowing sub-optimal usage patterns.
Activity
ohmayr commentedon Sep 12, 2026
System tests are passing on the individual release #18340. Given that to be the case, the release isn't blocked so it is reasonable to downgrade this to a P2. However, we still need to fix the flaky tests and determine we should add this back to the bulk release.
chalmerlowe commentedon Sep 21, 2026
The system test results do not display details of each test run. Merely which packages ran and the fact that the whole test timed out.
The packages that ran system tests in this session include these historically long-running sets of tests:
bigquerycomputespannerbigtableThis is a case of the session timed out as opposed to there was a specific flaky test.
We need to continue working to decrease the turn-around time for system tests across the board.
For the record, the spanner test in isolation takes ~31 minutes in toto(with setup/teardown)
system-3.12(protobuf_implementation='python', database_dialect='GOOGLE_STANDARD_SQL'): 15 minutessystem-3.12(protobuf_implementation='python', database_dialect='POSTGRESQL'): 11 minuteschalmerlowe commentedon Sep 21, 2026
Closing as OBE.
We are choosing to leave certain libraries out of bulk release to help control instances of CI timeouts and to grant product teams more flexibility over their own release schedule.