HDDS-13927. Recon - OM Db delta updates handler resource cleanup.

[devmadhuu]

What changes were proposed in this pull request?

This PR is to do cleanup of resources like rocksDB tables references from OMDbUpdatesHandler in case of any exception or error in following scenarios:

Specific Failure Scenarios When OMDbUpdatesHandler could still hold some references
Scenario 1: Exception During Batch Processing

try (OMDBUpdatesHandler omdbUpdatesHandler = new OMDBUpdatesHandler(omMetadataManager)) {

    for (byte[] data : dbUpdates.getData()) {

        try (ManagedWriteBatch writeBatch = new ManagedWriteBatch(data))

{             writeBatch.iterate(omdbUpdatesHandler);  // Handler stores references to RocksDB objects                          // EXCEPTION OCCURS HERE (e.g., RocksDBException during commit)             rdbBatchOperation.commit(rocksDB, wOpts);                        }
// writeBatch.close() called, but omdbUpdatesHandler still holds references!

    }

} // omdbUpdatesHandler "close" called, but it's a no-op since no close() method exists

Result: RocksDB handles, iterators, and memory remain allocated indefinitely.

Scenario 2: Interrupted Thread

try (OMDBUpdatesHandler omdbUpdatesHandler = new OMDBUpdatesHandler(omMetadataManager)) {

    // Thread gets interrupted during long-running operation

    if (Thread.currentThread().isInterrupted())

{         throw new InterruptedException("Thread interrupted during delta update.");     }
    // omdbUpdatesHandler has already accessed tables and created internal resources

    // Interrupt causes immediate exit without proper cleanup

}

What is the link to the Apache JIRA

https://issues.apache.org/jira/browse/HDDS-13927

How was this patch tested?

Tested the patch locally using docker cluster and existing tests.

[ArafatKhan2198]

Thanks for working on this @devmadhuu
Changes LGTM +1

[adoroszlai]

When merging PRs, please remove co-author information if it's the same person with different email address.