hadoop-yarn-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jason Lowe (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (YARN-8209) NPE in DeletionService
Date Tue, 01 May 2018 16:26:00 GMT

    [ https://issues.apache.org/jira/browse/YARN-8209?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16459819#comment-16459819
] 

Jason Lowe commented on YARN-8209:
----------------------------------

bq. In doing so, the exit code of docker overlaps with container-executor. This could confuse
node manager which reads exit code from container-executor.

run_docker is already doing execvp, so this behavior is not new with this patch.  In practice
this is not an issue since the container executor exit code is essentially treated a boolean
where zero means success and nonzero means failure.  docker (and just about every other Linux
command) has the same behavior, so I believe the exec is appropriate here.  I personally would
rather get the actual exit code from the docker inspect command logged in the NM logs rather
than an exit code that meant "docker failed in some way" that hides the specific exit code
from the docker command.


> NPE in DeletionService
> ----------------------
>
>                 Key: YARN-8209
>                 URL: https://issues.apache.org/jira/browse/YARN-8209
>             Project: Hadoop YARN
>          Issue Type: Bug
>            Reporter: Chandni Singh
>            Assignee: Eric Badger
>            Priority: Major
>         Attachments: YARN-8209.001.patch, YARN-8209.002.patch
>
>
> {code:java}
> 2018-04-25 23:38:41,039 WARN  concurrent.ExecutorHelper (ExecutorHelper.java:logThrowableFromAfterExecute(63))
- Caught exception in thread DeletionService #1:
> java.lang.NullPointerException
>         at org.apache.hadoop.yarn.server.nodemanager.containermanager.linux.runtime.docker.DockerClient.writeCommandToTempFile(DockerClient.java:109)
>         at org.apache.hadoop.yarn.server.nodemanager.containermanager.linux.runtime.docker.DockerCommandExecutor.executeDockerCommand(DockerCommandExecutor.java:85)
>         at org.apache.hadoop.yarn.server.nodemanager.containermanager.linux.runtime.docker.DockerCommandExecutor.executeStatusCommand(DockerCommandExecutor.java:192)
>         at org.apache.hadoop.yarn.server.nodemanager.containermanager.linux.runtime.docker.DockerCommandExecutor.getContainerStatus(DockerCommandExecutor.java:128)
>         at org.apache.hadoop.yarn.server.nodemanager.LinuxContainerExecutor.removeDockerContainer(LinuxContainerExecutor.java:935)
>         at org.apache.hadoop.yarn.server.nodemanager.containermanager.deletion.task.DockerContainerDeletionTask.run(DockerContainerDeletionTask.java:61)
>         at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
>         at java.util.concurrent.FutureTask.run(FutureTask.java:266)
>         at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$201(ScheduledThreadPoolExecutor.java:180)
>         at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:293)
>         at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
>         at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
>         at java.lang.Thread.run(Thread.java:748){code}



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: yarn-issues-help@hadoop.apache.org


Mime
View raw message