Github user vanzin commented on the pull request: https://github.com/apache/spark/pull/9021#issuecomment-146617647 > It's because we use a thread pool to call sendRpc. That too, but it can also happen if you enable multiple connections per peer (`spark.shuffle.io.numConnectionsPerPeer`). I still think for the RPC library we should hardcode that to `1` - that should also make implementing SPARK-10997 simpler. The bugs I saw in the transport library weren't really related to connection establishment, but I can take a look at how hard it would be to make connections non-blocking.
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. --- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org