Github user rdblue commented on the issue: https://github.com/apache/spark/pull/21558 @vanzin, the ID that this uses is the TID, which I thought was always unique. It appears to be a one-up counter. Also, I noted on your PR that both are needed because even if we only commit one of the attempts, the writers may use this ID to track and clean up data.
--- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org