Wenzhe Zhou has posted comments on this change. ( http://gerrit.cloudera.org:8080/21491 )
Change subject: IMPALA-13134 (part 2): Fix flaky test_catalogd_failover_with_sync_ddl ...................................................................... Patch Set 2: (4 comments) http://gerrit.cloudera.org:8080/#/c/21491/1/tests/custom_cluster/test_catalogd_ha.py File tests/custom_cluster/test_catalogd_ha.py: http://gerrit.cloudera.org:8080/#/c/21491/1/tests/custom_cluster/test_catalogd_ha.py@35 PS1, Line 35: _BUI > Is catalogd restart tend to slow? What factor in long catalogd restart time Doubled the base delay to 10s and 20s for slow build flavor. This happened for s3 nightly builds. Saw Joe fixed another flaky test on s3 due to long catalogd restart time during test. http://gerrit.cloudera.org:8080/#/c/21491/1/tests/custom_cluster/test_catalogd_ha.py@448 PS1, Line 448: e > flake8: E131 continuation line unaligned for hanging indent fixed http://gerrit.cloudera.org:8080/#/c/21491/1/tests/custom_cluster/test_catalogd_ha.py@470 PS1, Line 470: # Restart standby catalogd with force_catalogd_active as true. : start_s = time.time() > Can we log/assert the time until second catalogd active? Added log and assert http://gerrit.cloudera.org:8080/#/c/21491/1/tests/custom_cluster/test_catalogd_ha.py@475 PS1, Line 475: > Should this be adjusted too? No, since the restarted catalogd is already ready. -- To view, visit http://gerrit.cloudera.org:8080/21491 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: Impala-ASF Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I15bb6aae23a2f544067f993533e322969372ebd5 Gerrit-Change-Number: 21491 Gerrit-PatchSet: 2 Gerrit-Owner: Wenzhe Zhou <[email protected]> Gerrit-Reviewer: Impala Public Jenkins <[email protected]> Gerrit-Reviewer: Riza Suminto <[email protected]> Gerrit-Reviewer: Wenzhe Zhou <[email protected]> Gerrit-Comment-Date: Fri, 07 Jun 2024 17:30:32 +0000 Gerrit-HasComments: Yes
