[HDDS-3380] MiniOzoneHAClusterImpl#initOMRatisConf will reset the configs and causes for test failures

• Type: Bug
• Status: Resolved
• Priority: Minor
• Resolution: Fixed
• Affects Version/s: 0.5.0
• Fix Version/s: 1.0.0
• Component/s: OM HA, test
• Labels: pull-request-available
• Target Version/s: 1.0.0
• Assignee: Uma Maheswara Rao G
• Reporter: Uma Maheswara Rao G
• Created: 12/Apr/20 10:50
• Updated: 04/Jun/20 20:17
• Resolved: 21/Apr/20 16:34

Description

While I was debugging some code paths using miniOzoneCluster:

For example in TestOzoneHAManager:

it plans to trigger snapshots at threshold 50 and same was configured and passed to MiniOzoneHACluster. But inside MiniOzoneHAClusterImpl#initOMRatisConf, it will silently reset to 100L. So, test will expect snapshot to trigger after 50 transactions, but it will not.

 

It will keep wait even after rolling at 50:

GenericTestUtils.waitFor(() -> {
if (ozoneManager.getRatisSnapshotIndex() > 0)

Unknown macro: { return true; }

return false;
}, 1000, 100000);

 

2020-04-12 03:54:21,296 [omNode-1@group-523986131536-SegmentedRaftLogWorker] INFO segmented.SegmentedRaftLogWorker (SegmentedRaftLogWorker.java:execute(583)) - omNode-1@group-523986131536-SegmentedRaftLogWorker: created new log segment /Users/ugangumalla/Work/repos/hadoop-ozone/hadoop-ozone/integration-test/target/test-dir/MiniOzoneClusterImpl-fce544cd-3a80-4b0b-ac92-463cf391975c/omNode-1/ratis/c9bc4cf4-3bc3-3c60-a66b-523986131536/current/log_inprogress_49

 

So, respecting user passed configurations will fix the issue. I will post the patch later in some time.

Issue Links

• links to: GitHub Pull Request #817