Sorabh

@dtaivpp Checking again if you plan to clean this up ?

@sachinpkale Any update on adding tests around this change ?

@Gaganjuneja @reta Do we still plan to iterate on this PR (considering this is introducing a breaking change ?) If not, then can we consider closing this PR out ?

@msfroh Thanks for the discussion and proposal. As you called out both the approaches 1 & 2 are mutually exclusive and I think we will need both. I think option...

@kaushalmahi12 Agreed and that is what the issue is trying to explain. I think we should check the duress condition for each tracker as well. For example: If under heap...

Reopening this as again seeing this test failing: Ref CI: https://build.ci.opensearch.org/job/gradle-check/25984/ ``` REPRODUCE WITH: ./gradlew ':server:internalClusterTest' --tests "org.opensearch.snapshots.DeleteSnapshotIT.testDeleteShallowCopySnapshot" -Dtests.seed=5A77171FC14EEBF7 -Dtests.security.manager=true -Dtests.jvm.argline="-XX:TieredStopAtLevel=1 -XX:ReservedCodeCacheSize=64m" -Dtests.locale=vi -Dtests.timezone=PRC -Druntime.java=20 ``` ``` java.lang.AssertionError: Expected: is...

@harishbhakuni Can you take a look at this ?

Looking into the code around this area, seems like connection is always expected to be non-null (Ref [here](https://github.com/opensearch-project/OpenSearch/blob/main/server/src/main/java/org/opensearch/transport/TransportService.java#L939)). So probably adding a check `Objects.nonNull(connection, "Input connection is expected to be...

Another test is flaky in the same class: ``` REPRODUCE WITH: ./gradlew ':server:internalClusterTest' --tests "org.opensearch.remotestore.multipart.RemoteStoreMultipartIT.testAsyncDurabilityThrowsExceptionWhenRestrictSettingTrue" -Dtests.seed=5083B88EA28AD82D -Dtests.security.manager=true -Dtests.jvm.argline="-XX:TieredStopAtLevel=1 -XX:ReservedCodeCacheSize=64m" -Dtests.locale=el-POLYTON -Dtests.timezone=America/Adak -Druntime.java=21 com.carrotsearch.randomizedtesting.UncaughtExceptionError: Captured an uncaught exception in thread: Thread[id=4818,...

@jpountz This looks interesting. On similar lines to this, in OpenSearch we are working on building a warm index which will not have all the data available locally all the...