Marc Handalian

@dreamer-89 I prefer option 1 of isolating these tests to a separate task where we pass the jvm flag, as long as they are included with check & run in...

Sorry for the no reply here @tan31989. To test I would suggest creating a test feature flag that is passed in an IT as a node level setting with `internalCluster().startNode(nodeSettings);`....

@tan31989 Do you still have interest in picking this up? If not @techrajdeep its all yours.

Thanks @tan31989 I will assign to you. Let me know if you have further questions on the issue.

Thanks for your interest @techrajdeep! I would suggest filtering by the tag [good first issue](https://github.com/opensearch-project/OpenSearch/issues?q=is%3Aopen+is%3Aissue+label%3A%22good+first+issue%22) to find these issues.

Re-opeening, failed with https://github.com/opensearch-project/OpenSearch/pull/12877

Have left this running overnight on main - pre #9480 merge and not seeing this after ~12k iterations. Will try pulling in latest changes. We likely have a race closing...

> This issue is not gone or fixed: #13446 @reta I believe this is a diff test in https://build.ci.opensearch.org/job/gradle-check/37929/ [SegmentReplicationIT.testReplicaAlreadyAtCheckpoint](https://build.ci.opensearch.org/job/gradle-check/37929/testReport/junit/org.opensearch.indices.replication/SegmentReplicationIT/testReplicaAlreadyAtCheckpoint/) which is a diff test suite - opened issue https://github.com/opensearch-project/OpenSearch/issues/13593...

@linuxpi I have seen this as well but not able to get a repeatable test. In any case I don't think we need the engineMutex while `getSegmentInfosSnapshot` is invoked on...

Grr, hit a failure here locally after about ~1k more iterations for the remote store version of this test where segments are still part of the latest local commit on...