juliuszsompolski commented on a change in pull request #25062:
[SPARK-28260][SQL] Add CLOSED state to ExecutionState
URL: https://github.com/apache/spark/pull/25062#discussion_r301047128
##########
File path:
sql/hive-thriftserver/src/main/scala/org/apache/spark/sql/hive/thriftserver/HiveThriftServer2.scala
##########
@@ -254,6 +254,11 @@ object HiveThriftServer2 extends Logging {
trimExecutionIfNecessary()
}
+ def onOperationClosed(id: String): Unit = synchronized {
+ executionList(id).finishTimestamp = System.currentTimeMillis
Review comment:
I would vote for the former - add `Close time` and `Execution time` and
`Duration`.
`Fetch result time` is a bit of a long label, and also I think it might be
slightly misleading - this may also be the client being slow in processing the
results, just sitting on an idle cursor, or actually closing it without
fetching all results etc.
Having `Duration` for the total time gives all the same information and I
think it's also more relevant, to have the time from start to close.
----------------------------------------------------------------
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]
