[
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-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
