JobQueues-LongRunningJobs


#1

I have observed that at times some users queries/codes sumitted on the cluster are so long that they are allocated high number of containers and take very long time to execute due to which jobs of other users have to wait very long. My query for the past 30 minutes is in wait.
Cannot there be any mechanism for aborting jobs taking too long time.
Else the lab will loose it’s purpose of enabling the users to test the hadoop basics.

I have attached a snap.


Please do the needful.