-
Type:
Sub-task
-
Status: Resolved
-
Priority:
Major
-
Resolution: Fixed
-
Affects Version/s: None
-
Fix Version/s: YARN-1011
-
Component/s: nodemanager
-
Labels:None
-
Hadoop Flags:Reviewed
As discussed in YARN-8250, we can stop opportunistic containers from being launched at container scheduler/finish events if the node is already over-allocating itself. This can mitigate the issue that too many opportunistic containers can be launched and then quickly killed.