DanielLeens opened a new pull request, #11180:
URL: https://github.com/apache/seatunnel/pull/11180
## What does this PR do?
Fixes a shared SFTP cleanup failure that showed up in unrelated PR CI runs.
The fix keeps pooled sibling channels tracked correctly, closes stale SSH
sessions when a dead channel is replaced, and initializes the pool with the
real live-connection count so reusable sessions are not treated as over-limit
immediately.
## Why is this needed?
Recent CI failures in
`SftpFileIT#testSftpBinaryUpdateModeContinuousDiscoveryDistcp` were not caused
by the feature PRs under test. The failing job reported `Wait continuous job
exit failed` and `There are still threads running in the container`, which
matches leaked or stale SFTP session cleanup.
## How was this validated?
- `./mvnw -pl seatunnel-connectors-v2/connector-file/connector-file-sftp
spotless:apply -DskipTests`
- `./mvnw -pl seatunnel-connectors-v2/connector-file/connector-file-sftp
-Dtest=SFTPConnectionPoolTest,SftpFileSystemTest -DfailIfNoTests=false test`
- attempted closer E2E validation with:
- `./mvnw -pl
seatunnel-e2e/seatunnel-connector-v2-e2e/connector-file-sftp-e2e -Dskip.ui=true
-Dtest=SftpFileIT#testSftpBinaryUpdateModeContinuousDiscoveryDistcp
-DfailIfNoTests=false test`
`seatunnel-engine-ui` was skipped because this fix only touches the SFTP
connector module and the SFTP E2E module. The E2E attempt was blocked locally
by an unresponsive Docker socket (`/_ping` timed out on
`~/.docker/run/docker.sock`), so the closest completed validation is the
connector module unit test coverage above.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
