Github user jose-torres commented on the issue:
https://github.com/apache/spark/pull/20622
The alternate solution doesn't solve the problem SPARK-23441 sets out to
address. Using a thread interrupt to kill the spark job is a fundamentally
fragile solution, requiring us to maintain a whitelist of exceptions we think
Spark execution might surface in response to an interrupt.
If you don't think there's a simple way to remove this interrupt currently,
I can close this PR and write a new one that simply addresses the test
flakiness using the alternate solution.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
