[HDDS-6470] Fix TestOzoneManagerHAWithData#testOMRestart()

• Type: Sub-task
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 1.3.0
• Component/s: test
• Labels: pull-request-available
• Assignee: Kaijie Chen
• Reporter: Kaijie Chen
• Created: 19/Mar/22 05:45
• Updated: 26/Apr/22 02:55
• Resolved: 19/Mar/22 09:17

Description

This assertion is wrong in TestOzoneManagerHAWithData#testOMRestart().

Because the lagging follower OM may catch up asynchronously.

    // Restart the stopped OM.
    followerOM1.restart();
 
    // Get the latest snapshotIndex from the leader OM.
    long leaderOMSnaphsotIndex = leaderOM.getRatisSnapshotIndex();
 
    // The recently started OM should be lagging behind the leader OM.
    long followerOMLastAppliedIndex =
        followerOM1.getOmRatisServer().getLastAppliedTermIndex().getIndex();
    Assert.assertTrue(
        followerOMLastAppliedIndex < leaderOMSnaphsotIndex);

Example of CI failure on master: https://github.com/apache/ozone/runs/5593803014

Result of running this test 100x: https://github.com/kaijchen/ozone/actions/runs/2007487998

Issue Links

• duplicates: HDDS-6469 Intermittent failure in TestOzoneManagerHAWithData#testOMRestart
• links to: GitHub Pull Request #3213