[
https://issues.apache.org/jira/browse/YARN-6570?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16001661#comment-16001661
]
Junping Du commented on YARN-6570:
----------------------------------
Look at this issue together with [[email protected]] offline. The issue
is that the container didn't get actually running by the time the test invoked
the command.
The container status reported as RUNNING from RM side is earlier than container
status get actually running on NM:
{noformat}
vim vim
./hadoop-yarn-project/hadoop-yarn/hadoop-yarn-server/hadoop-yarn-server-nodemanager/src/main/java/org/apache/hadoop/yarn/server/nodemanager/containermanager/container/ContainerImpl.java
...
case NEW:
case LOCALIZING:
case LOCALIZATION_FAILED:
case LOCALIZED:
case RUNNING:
case EXITED_WITH_SUCCESS:
case EXITED_WITH_FAILURE:
case KILLING:
case CONTAINER_CLEANEDUP_AFTER_KILL:
case CONTAINER_RESOURCES_CLEANINGUP:
return org.apache.hadoop.yarn.api.records.ContainerState.RUNNING;
{noformat}
We should fix it by using ContainerState.NEW to represent states before
actually RUNNING.
> No logs were found for running application, running container
> -------------------------------------------------------------
>
> Key: YARN-6570
> URL: https://issues.apache.org/jira/browse/YARN-6570
> Project: Hadoop YARN
> Issue Type: Bug
> Components: yarn
> Reporter: Sumana Sathish
> Assignee: Junping Du
> Priority: Critical
>
> 1.Obtain running containers from the following CLI for running application:
> yarn container -list appattempt
> 2. Couldnot fetch logs
> {code}
> Can not find any log file matching the pattern: ALL for the container
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
