Use SegmentReplicationTarget test to validate non-active on-disk files are reused for replication

[dreamer-89]

Description

This change updates testSegRepSucceedsOnPreviousCopiedFiles shard level test to wait for file copy operations to complete in first round of segment replication before proceeding to next round without which it is possible to have concurrent access on segment files. 1) FIle copy operations from first round 2) File checksum validation from next round.

Issues Resolves

Resolves #10885

Background

With #9630, we introduced optimization to avoid re-downloading files already on disk but not referenced by active reader. With #10725, we made RemoteStoreReplicationSource compliant with ReplicationSource by making download logic asynchronous. I suspect this resulted scenario where file copy remains in-progress from first round of segment replication which breaks checksum validations on next round.

Error trace for failure builds

Caused by: java.io.UncheckedIOException: Error reading name [_0.si], length [325], checksum [r5w8xw], writtenBy [9.8.0]
	at org.opensearch.indices.replication.SegmentReplicationTarget.validateLocalChecksum(SegmentReplicationTarget.java:246) 
...
Caused by: java.nio.file.AccessDeniedException: MockDirectoryWrapper: file "_0.si" is still open for writing
	at org.apache.lucene.tests.store.MockDirectoryWrapper.openInput(MockDirectoryWrapper.java:808) ~[lucene-test-framework-9.8.0.jar:9.8.0 d914b3722bd5b8ef31ccf7e8ddc638a87fd648db - 2023-09-21 21:57:47]
	at org.apache.lucene.store.FilterDirectory.openInput(FilterDirectory.java:101) ~[lucene-core-9.8.0.jar:9.8.0 d914b3722bd5b8ef31ccf7e8ddc638a87fd648db - 2023-09-21 21:57:47]
	at org.apache.lucene.store.FilterDirectory.openInput(FilterDirectory.java:101) ~[lucene-core-9.8.0.jar:9.8.0 d914b3722bd5b8ef31ccf7e8ddc638a87fd648db - 2023-09-21 21:57:47]
	at org.opensearch.indices.replication.SegmentReplicationTarget.validateLocalChecksum(SegmentReplicationTarget.java:236) ~[main/:?]
...
Caused by: java.lang.RuntimeException: unclosed IndexOutput: _0.si
	at org.apache.lucene.tests.store.MockDirectoryWrapper.addFileHandle(MockDirectoryWrapper.java:783) ~[lucene-test-framework-9.8.0.jar:9.8.0 d914b3722bd5b8ef31ccf7e8ddc638a87fd648db - 2023-09-21 21:57:47]
	at org.apache.lucene.tests.store.MockDirectoryWrapper.createOutput(MockDirectoryWrapper.java:719) ~[lucene-test-framework-9.8.0.jar:9.8.0 d914b3722bd5b8ef31ccf7e8ddc638a87fd648db - 2023-09-21 21:57:47]
	at org.apache.lucene.store.FilterDirectory.createOutput(FilterDirectory.java:75) ~[lucene-core-9.8.0.jar:9.8.0 d914b3722bd5b8ef31ccf7e8ddc638a87fd648db - 2023-09-21 21:57:47]
	at org.opensearch.index.store.ByteSizeCachingDirectory.createOutput(ByteSizeCachingDirectory.java:153) ~[main/:?]
        at org.apache.lucene.store.Directory.copyFrom(Directory.java:181) ~[lucene-core-9.8.0.jar:9.8.0 d914b3722bd5b8ef31ccf7e8ddc638a87fd648db - 2023-09-21 21:57:47]
	at org.opensearch.index.store.Store$StoreDirectory.copyFrom(Store.java:978) ~[main/:?]

testSegRepSucceedsOnPreviousCopiedFiles steps

Step 1. Index docs
Step 2. Use custom RemoteStoreReplicationSource where exception is post getSegmentFiles. With background files are been copied, we halted first round of segment replication pre-maturely. Below code shows the custom remote store replication source used during first round of segrep. We throw exception post getsegmentFiles method call.

        return new RemoteStoreReplicationSource(shard) {
            @Override
            public void getCheckpointMetadata(
                long replicationId,
                ReplicationCheckpoint checkpoint,
                ActionListener<CheckpointInfoResponse> listener
            ) {
                super.getCheckpointMetadata(replicationId, checkpoint, listener);
            }

            @Override
            public void getSegmentFiles(
                long replicationId,
                ReplicationCheckpoint checkpoint,
                List<StoreFileMetadata> filesToFetch,
                IndexShard indexShard,
                BiConsumer<String, Long> fileProgressTracker,
                ActionListener<GetSegmentFilesResponse> listener
            ) {
                super.getSegmentFiles(replicationId, checkpoint, filesToFetch, indexShard, (fileName, bytesRecovered) -> {}, listener);
                postGetFilesRunnable.run(); --> This can be invoked while files download is in progress, which fiddles with next round of segment replication validateCheckSum method call.
            }

3. Validate on-disk store is non-empty (validates previous step).
4. Perform next round of segment replication and validate on-disk files are considered.
   Issue: At step 2, since download (getSegmentFiles) is async in nature, throwing exception right after getSegmentFiles is not right. We need mechanism that tells files were downloaded completely.

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]

❌ Gradle check result for 8b69e04: 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?

[github-actions]

Compatibility status:

Checks if related components are compatible with change 95a866a

Incompatible components

Skipped components

Compatible components

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

[github-actions]

✅ Gradle check result for 95a866a: SUCCESS

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (cbfe160) 71.31% compared to head (95a866a) 71.18%.
Report is 3 commits behind head on main.

Additional details and impacted files

@@             Coverage Diff              @@
##               main   #11786      +/-   ##
============================================
- Coverage     71.31%   71.18%   -0.14%     
+ Complexity    59221    59149      -72     
============================================
  Files          4911     4911              
  Lines        278667   278667              
  Branches      40528    40528              
============================================
- Hits         198735   198363     -372     
- Misses        63375    63776     +401     
+ Partials      16557    16528      -29     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.