srowen commented on pull request #28769: URL: https://github.com/apache/spark/pull/28769#issuecomment-641655314
I'm not sure, I don't see any usages of it that don't close the iterator in a finally block. I do wonder where these are allocated, if there is a straightforward fix to the resource leak elsewhere. I don't see a scenario where the DB can close before the iterator does, except maybe shutdown, not sure. If it's reproducible, one interesting hack you could put in here is to record the stack trace where the iterator is constructed temporarily, then see what's left to close in your new code path, to debug where the leaked ones came from? ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
