[Snapshot Interop][Bug Fix] Ensure lock file gets deleted if snapshot …

[harishbhakuni]

...shard md upload fails during snapshot creation.

Description

• While working on another change, realized that in the following code, everything below snapshotRemoteStoreIndexShard call was not getting executed:
  

  OpenSearch/server/src/main/java/org/opensearch/snapshots/SnapshotShardsService.java

  Lines 427 to 462 in 829215c

  	repository.snapshotRemoteStoreIndexShard(
  	indexShard.store(),
  	snapshot.getSnapshotId(),
  	indexId,
  	snapshotIndexCommit,
  	getShardStateId(indexShard, snapshotIndexCommit),
  	snapshotStatus,
  	primaryTerm,
  	startTime,
  	ActionListener.runBefore(listener, wrappedSnapshot::close)
  	);
  	} catch (IndexShardSnapshotFailedException e) {
  	logger.error(
  	"Shallow Copy Snapshot Failed for Shard ["
  	+ indexId.getName()
  	+ "]["
  	+ shardId.getId()
  	+ "] for snapshot "
  	+ snapshot.getSnapshotId()
  	+ ", releasing acquired lock from remote store"
  	);
  	indexShard.releaseLockOnCommitData(snapshot.getSnapshotId().getUUID(), primaryTerm, commitGeneration);
  	throw e;
  	}
  	long endTime = threadPool.relativeTimeInMillis();
  	logger.debug(
  	"Time taken (in milliseconds) to complete shallow copy snapshot, "
  	+ "for index "
  	+ indexId.getName()
  	+ ", shard "
  	+ shardId.getId()
  	+ " and snapshot "
  	+ snapshot.getSnapshotId()
  	+ " is "
  	+ (endTime - startTime)
  	);

• that was causing an issue where if shard md upload to snapshot repository fails, it will not release the lock file from S3.

• as part of this PR, making sure those lines of code get executed even if shard md upload fails and adding IT to cover that scenario.

• Also, Adding a log line to print the time taken in flush operation during snapshot creation and another minor change to filter lock files based on lock suffix while fetching lock files from lock directory.

Check List

• New functionality includes testing.
  • All tests pass
• New functionality has been documented.
  • New functionality has javadoc added
• Failing checks are inspected and point to the corresponding known issue(s) (See: Troubleshooting Failing Builds)
• Commits are signed per the DCO using --signoff
• Commit changes are listed out in CHANGELOG.md file (See: Changelog)
• Public documentation issue/PR created

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.

[github-actions]

Compatibility status:

Checks if related components are compatible with change 197ccfd

Incompatible components

Incompatible components: [https://github.com/opensearch-project/cross-cluster-replication.git, https://github.com/opensearch-project/performance-analyzer.git, https://github.com/opensearch-project/performance-analyzer-rca.git]

Skipped components

Compatible components

Compatible components: [https://github.com/opensearch-project/security-analytics.git, https://github.com/opensearch-project/geospatial.git, https://github.com/opensearch-project/observability.git, https://github.com/opensearch-project/notifications.git, https://github.com/opensearch-project/opensearch-oci-object-storage.git, https://github.com/opensearch-project/job-scheduler.git, https://github.com/opensearch-project/custom-codecs.git, https://github.com/opensearch-project/neural-search.git, https://github.com/opensearch-project/security.git, https://github.com/opensearch-project/asynchronous-search.git, https://github.com/opensearch-project/reporting.git, https://github.com/opensearch-project/anomaly-detection.git, https://github.com/opensearch-project/ml-commons.git, https://github.com/opensearch-project/common-utils.git, https://github.com/opensearch-project/sql.git, https://github.com/opensearch-project/index-management.git, https://github.com/opensearch-project/alerting.git, https://github.com/opensearch-project/k-nn.git]

[github-actions]

❌ Gradle check result for b8ec898: FAILURE

Please examine the workflow log, locate, and copy-paste the failure(s) below, then iterate to green. Is the failure a flaky test unrelated to your change?

[harishbhakuni]

Test org.opensearch.search.sort.FieldSortIT.testSimpleSorts {p0={"search.concurrent_segment_search.enabled":"true"}} is failing which is flaky and being tracked as part of #11875 .

[opensearch-trigger-bot]

This PR is stalled because it has been open for 30 days with no activity.

[github-actions]

❌ Gradle check result for 4cdd33e: FAILURE

Please examine the workflow log, locate, and copy-paste the failure(s) below, then iterate to green. Is the failure a flaky test unrelated to your change?

[harishbhakuni]

org.opensearch.cluster.coordination.AwarenessAttributeDecommissionIT.testConcurrentDecommissionAction is flaky: #12197

[github-actions]

❌ Gradle check result for 203e404: FAILURE

Please examine the workflow log, locate, and copy-paste the failure(s) below, then iterate to green. Is the failure a flaky test unrelated to your change?

[harishbhakuni]

#13600
org.opensearch.indices.IndicesRequestCacheIT.testDeleteAndCreateSameIndexShardOnSameNode {p0={"search.concurrent_segment_search.enabled":"false"}},
org.opensearch.indices.IndicesRequestCacheIT.testDeleteAndCreateSameIndexShardOnSameNode {p0={"opensearch.experimental.feature.pluggable.caching.enabled":"true"}},
org.opensearch.indices.IndicesRequestCacheIT.testDeleteAndCreateSameIndexShardOnSameNode {p0={"search.concurrent_segment_search.enabled":"true"}},
org.opensearch.indices.IndicesRequestCacheIT.testDeleteAndCreateSameIndexShardOnSameNode {p0={"opensearch.experimental.feature.pluggable.caching.enabled":"false"}}

#13939
org.opensearch.remotemigration.RemoteMigrationIndexMetadataUpdateIT.testRemoteIndexPathFileExistsAfterMigration:

[github-actions]

❌ Gradle check result for 184132e: FAILURE

Please examine the workflow log, locate, and copy-paste the failure(s) below, then iterate to green. Is the failure a flaky test unrelated to your change?

[harishbhakuni]

org.opensearch.indices.IndicesRequestCacheIT.classMethod: test suite timed out.

All the below mentioned tests are flaky:
org.opensearch.action.admin.indices.create.RemoteCloneIndexIT.testCreateCloneIndexFailure: #13476
org.opensearch.cluster.MinimumClusterManagerNodesIT.testThreeNodesNoClusterManagerBlock: #10006
org.opensearch.indices.IndicesRequestCacheIT.testStaleKeysCleanupWithMultipleIndices {p0={"opensearch.experimental.feature.pluggable.caching.enabled":"false"}}
org.opensearch.indices.IndicesRequestCacheIT.classMethod: #13437

[github-actions]

❌ Gradle check result for 1ebbc06: FAILURE

Please examine the workflow log, locate, and copy-paste the failure(s) below, then iterate to green. Is the failure a flaky test unrelated to your change?

[harishbhakuni]

org.opensearch.action.admin.indices.create.RemoteCloneIndexIT.testCreateCloneIndexFailure: #13476
org.opensearch.indices.IndicesRequestCacheIT.testCacheCleanupWithDefaultSettings {p0={"opensearch.experimental.feature.pluggable.caching.enabled":"true"}}
org.opensearch.indices.IndicesRequestCacheIT.classMethod: #13711

[github-actions]

❌ Gradle check result for 57799ac: FAILURE

Please examine the workflow log, locate, and copy-paste the failure(s) below, then iterate to green. Is the failure a flaky test unrelated to your change?

[gbbafna]

"that was causing an issue where if shard md upload to snapshot repository fails, it will not release the lock file from S3." - How is that happening ?

OpenSearch/server/src/main/java/org/opensearch/snapshots/SnapshotShardsService.java

Lines 438 to 462 in 829215c

	} catch (IndexShardSnapshotFailedException e) {
	logger.error(
	"Shallow Copy Snapshot Failed for Shard ["
	+ indexId.getName()
	+ "]["
	+ shardId.getId()
	+ "] for snapshot "
	+ snapshot.getSnapshotId()
	+ ", releasing acquired lock from remote store"
	);
	indexShard.releaseLockOnCommitData(snapshot.getSnapshotId().getUUID(), primaryTerm, commitGeneration);
	throw e;
	}
	long endTime = threadPool.relativeTimeInMillis();
	logger.debug(
	"Time taken (in milliseconds) to complete shallow copy snapshot, "
	+ "for index "
	+ indexId.getName()
	+ ", shard "
	+ shardId.getId()
	+ " and snapshot "
	+ snapshot.getSnapshotId()
	+ " is "
	+ (endTime - startTime)
	);

The above block is doing exactly that

[harishbhakuni]

The above block is doing exactly that

not really, the problem was that caller method snapshotRemoteStoreIndexShard was catching the exception and passing it with listener.onFailure() method. so the catch block where we were releasing lock will never get executed. just moved that code to onFailure() method. so that it will get executed.

[gbbafna]

we should throw IndexShardSnapshotFailedException along with listener.onFailure(e);

OpenSearch/server/src/main/java/org/opensearch/repositories/blobstore/BlobStoreRepository.java

Line 2691 in 829215c

listener.onFailure(e);

[gbbafna]

nit : lets not use prod word . Also a user can create extra files in the lock folder as well