warrenzhu25 commented on PR #41083: URL: https://github.com/apache/spark/pull/41083#issuecomment-1548009051
> How are you observing recoverable fetch failures ? I have seen 2 cases when target executor has busy shuffle fetch and upload due to shuffle migration: 1. All Netty request handler threads are exhausted, it'll throw `TimeoutException` when creating connection to target executor took longer than `connectTimeout`. 2. `IdleStateHandler` will close connection when fetch request took longer than `requestTimeout` to receive response. Both cases are recoverable. -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
