[
https://issues.apache.org/jira/browse/YARN-4947?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15256194#comment-15256194
]
Rohith Sharma K S commented on YARN-4947:
-----------------------------------------
Thanks [~bibinchundatt] for the updating patch.
some comments
# I think {{rm.NMwaitForState(nm2.getNodeId(), NodeState.NEW);}} code can be
removed. I don't see any value add from this. For the test case
{{testNodesQueryNew}} you can hack it by creating new RMNodeImpl and add to
context explicitly. This will ensure, RMNode is in new state.
# If we handle 1st point, then MockRM changes do not required.
# {{rm.sendNodeStarted(nm1);}} is not required since RM is started which makes
RMNode to transition to running.
# At last, to avoid problems wrong usage an API, I think it is better to add
MockRM service started check in {{MockRM#registerNode}}
> Test timeout is happening for TestRMWebServicesNodes
> ----------------------------------------------------
>
> Key: YARN-4947
> URL: https://issues.apache.org/jira/browse/YARN-4947
> Project: Hadoop YARN
> Issue Type: Sub-task
> Reporter: Bibin A Chundatt
> Assignee: Bibin A Chundatt
> Attachments: 0001-YARN-4947.patch, 0002-YARN-4947.patch,
> 0003-YARN-4947.patch, 0004-YARN-4947.patch
>
>
> Testcase timeout for TestRMWebServicesNodes is happening after YARN-4893
> [timeout|https://builds.apache.org/job/PreCommit-YARN-Build/11044/testReport/]
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
