cloud-fan commented on issue #24467: [SPARK-27568][CORE] Fix readLock leak while calling take()/first() on a cached rdd URL: https://github.com/apache/spark/pull/24467#issuecomment-489882706 This reminds me of the memory leak issue in sort-merge-join. We use scala `Iterator` to exchange data between SQL operators/RDDs, and we can only do the cleanup work when the `Iterator` is consumed up, or in a task completion listener. But what we really need is a traditional database iterator, with `open`, `next`, `get` and `close` methods. When a downstream operator finishes its work without consuming up the input data, it can call `close` to do free the resource. For this particular case, I think using a task completion listener is good enough?
---------------------------------------------------------------- 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] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
