-
Notifications
You must be signed in to change notification settings - Fork 424
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
TEZ-4554: Counter for used nodes within a DAG #362
Conversation
This comment was marked as outdated.
This comment was marked as outdated.
This comment was marked as outdated.
This comment was marked as outdated.
This comment was marked as outdated.
This comment was marked as outdated.
This comment was marked as outdated.
This comment was marked as outdated.
This comment was marked as outdated.
This comment was marked as outdated.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
public void addUsedContainer(Container container) { | ||
containersUsedByCurrentDAG.add(container.getId()); | ||
nodesUsedByCurrentDAG.add(container.getNodeId()); | ||
nodeHostsUsedByCurrentDAG.add(container.getNodeId().getHost()); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
can container.getNodeId()
be null
in any scenario?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
hm, should not be, in that case it's a yarn bug
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ohhk
thanks @ayushtkn for the review |
This comment was marked as outdated.
This comment was marked as outdated.
thanks @ayushtkn for the review so far seems like I have to slightly change this patch before moving to HIVE-28201, because it returns all nodes by TaskSchedulerManager.getNumClusterNodes, which is served by a cachedNodeCount, local to the TaskSchedulerManager, it won't be accessible from the LlapTaskSchedulerService best scenario is that HIVE-28201 won't need any implementation I'll let you know later |
This comment was marked as outdated.
This comment was marked as outdated.
This comment was marked as outdated.
This comment was marked as outdated.
🎊 +1 overall
This message was automatically generated. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
tested on a small tez container mode cluster
with 2 healthy and running yarn nodemanagers:
after stopping 1 nodemanager:
after decommissioning 1 nodemanager:
also tested with LLAP on Cloudera CDW (3 LLAP daemons)