Problem Statement

When onboarding the blob transfer bootstrap feature to a large store (e.g., 10GB per partition, 120GB per host), the transfer time is so long that it triggers a client-side timeout exception. Upon reaching the timeout, a partition cleanup is performed before moving to the next host.

However, during the cleanup process, the channels are not closed, and Netty continues receiving transferred files. If files are being cleaned up while validation is happening, checksum failures occur, resulting in checksum errors. These failures trigger the exceptionCaught method, which eventually leads to the channel being closed.

As a result, incomplete cleanups occur—some files are deleted, but others that are still being transferred or created after the cleanup begins remain. This race condition arises because file transfers and cleanups are happening concurrently.

Ultimately, even if the blob transfer fails and the bootstrap falls back to Kafka ingestion, the incomplete cleanup leads to database corruption due to residual files.

Solution

1. Allow the client-side timeout to be configurable. Previously, we only have server-side timeout config.
2. Close the channel upon timeout to prevent continued file reception.
3. Reduce the server-side active user count when the channel becomes inactive. This prevents the server from maintaining an active connection count if the connection is unexpectedly interrupted.

Code changes

• Added new code behind a config. If so list the config names and their default values in the PR description.
• Introduced new log lines.
  • Confirmed if logs need to be rate limited to avoid excessive logging.

Concurrency-Specific Checks

Both reviewer and PR author to verify

• Code has no race conditions or thread safety issues.
• Proper synchronization mechanisms (e.g., synchronized, RWLock) are used where needed.
• No blocking calls inside critical sections that could lead to deadlocks or performance degradation.
• Verified thread-safe collections are used (e.g., ConcurrentHashMap, CopyOnWriteArrayList).
• Validated proper exception handling in multi-threaded code to avoid silent thread termination.

How was this PR tested?

1. Verify the reduction of the server-side active user count through integration tests and unit test
2. The timeout should be tested at the host E2E level, as integration tests complete too quickly (in seconds) to reach a timeout set at the minute level.

• New unit tests added.
• New integration tests added.
• Modified or extended existing tests.
• Verified backward compatibility (if applicable).

Does this PR introduce any user-facing or breaking changes?

• No. You can skip the rest of this section.
• Yes. Clearly explain the behavior change and its impact.