Wangda Tan commented on YARN-2494:

Hi [~cwelch],
Thanks for your comments,
bq. the only change is the (?automated) removal of an import, I think you 
should just drop it from the change set.
Good catch, reverted this file.

bq. why force all to lower case? Discussion favored dropping that...
Updated according to our discussion

bq. checks for valid labels, there must be an easier way/stringlib/regex
Good suggestion, updated

bq. also in updateLableResource - it looks like if node1 has label a b and 
queue q1 has label a b it’s resources will be added 2x and removed 2x, while 
present it will have a 2x value (1x too many)
It should not, you can check the test: 
{{TestNodeLabelManager#testGetQueueResource}}, it should cover the case you 

bq. line 603 exception message needs to include “or not present”
I found there's no exception msg around line 603, could you please update 
according to latest patch?

bq. pls rename activeNode deactiveNode to activateNode and deactivateNode

Attached a new patch address your comments, please kindly review.


> [YARN-796] Node label manager API and storage implementations
> -------------------------------------------------------------
>                 Key: YARN-2494
>                 URL: https://issues.apache.org/jira/browse/YARN-2494
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: resourcemanager
>            Reporter: Wangda Tan
>            Assignee: Wangda Tan
>         Attachments: YARN-2494.patch, YARN-2494.patch, YARN-2494.patch, 
> YARN-2494.patch
> This JIRA includes APIs and storage implementations of node label manager,
> NodeLabelManager is an abstract class used to manage labels of nodes in the 
> cluster, it has APIs to query/modify
> - Nodes according to given label
> - Labels according to given hostname
> - Add/remove labels
> - Set labels of nodes in the cluster
> - Persist/recover changes of labels/labels-on-nodes to/from storage
> And it has two implementations to store modifications
> - Memory based storage: It will not persist changes, so all labels will be 
> lost when RM restart
> - FileSystem based storage: It will persist/recover to/from FileSystem (like 
> HDFS), and all labels and labels-on-nodes will be recovered upon RM restart

This message was sent by Atlassian JIRA

Reply via email to