[SPARK-45245] PythonWorkerFactory create gets stuck in case of worker failures

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 4.0.0
• Fix Version/s: 4.0.0
• Component/s: PySpark
• Labels: pull-request-available
• Assignee: Raghu Angadi
• Reporter: Raghu Angadi
• Created: 21/Sep/23 04:00
• Updated: 11/Mar/24 18:00
• Resolved: 30/Oct/23 23:50

Description

`createSimpleWorker()` method in `PythonWorkerFactory` does not return if the worker fails to connect back to the server. 

This is because it calls accept() without a timeout. If the worker does not connect back, accept() waits forever. 

See PR for the fix. 

Issue Links

• links to: GitHub Pull Request #43023
• links to: GitHub Pull Request #45459