[
https://issues.apache.org/jira/browse/YARN-1273?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13786922#comment-13786922
]
Vinod Kumar Vavilapalli commented on YARN-1273:
-----------------------------------------------
Yup. The patch looks fine. I think we can easily write a test by directly using
ApplicationMaster and intercepting NM Client. Basically initialize
ApplicationMaster, add a new protected method for creating NMCallbackHandler,
intercept onStartContainerSuccess and call back onStartContainerError. Let me
know if that works.
> Distributed shell does not account for start container failures reported
> asynchronously.
> ----------------------------------------------------------------------------------------
>
> Key: YARN-1273
> URL: https://issues.apache.org/jira/browse/YARN-1273
> Project: Hadoop YARN
> Issue Type: Bug
> Reporter: Hitesh Shah
> Assignee: Hitesh Shah
> Attachments: YARN-1273.1_1.patch
>
>
> 2013-10-04 22:09:15,234 ERROR
> [org.apache.hadoop.yarn.client.api.async.impl.NMClientAsyncImpl #1]
> distributedshell.ApplicationMaster
> (ApplicationMaster.java:onStartContainerError(719)) - Failed to start
> Container container_1380920347574_0018_01_000006
--
This message was sent by Atlassian JIRA
(v6.1#6144)
