Siyao Meng created HDDS-17130:
---------------------------------
Summary: MiniOzoneHAClusterImpl#bootstrapSCM does not wait for the
bootstrapped SCM
Key: HDDS-17130
URL: https://issues.apache.org/jira/browse/HDDS-17130
Project: Apache Ozone
Issue Type: Sub-task
Components: test
Reporter: Siyao Meng
In {{MiniOzoneHAClusterImpl#bootstrapSCM}} the calls to
{{waitForBootstrappedNodeToBeReady}} and {{waitForConfigUpdateOnActiveSCMs}}
are inside the retry loop, after the {{break}} on success. They only run on the
BindException retry path, where {{scm}} may still be null, and never after a
successful bootstrap. Compare {{bootstrapOzoneManager}}, which waits after the
loop.
Also, {{waitForConfigUpdateOnActiveSCMs}} checks
{{scm.doesPeerExist(scm.getScmId())}}, i.e. whether each SCM has itself as a
peer, instead of whether it has the new SCM.
Tests using it (e.g. TestDatanodeSCMNodesReconfiguration) therefore do not wait
for the new SCM to catch up, which is a flakiness risk.
*Fix:* move both waits after the loop, and check
{{scm.doesPeerExist(newSCMNode.getScmId())}}.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
