You are viewing a plain text version of this content. The canonical link for it is here.
Posted to yarn-issues@hadoop.apache.org by "Feng Yuan (JIRA)" <ji...@apache.org> on 2017/05/26 03:12:04 UTC
[jira] [Comment Edited] (YARN-6630) Container worker dir could not
recover when NM restart
[ https://issues.apache.org/jira/browse/YARN-6630?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16025741#comment-16025741 ]
Feng Yuan edited comment on YARN-6630 at 5/26/17 3:11 AM:
----------------------------------------------------------
Hi, wy {code}ContainerRetryPolicy{code} is configuarable,for example if you are using DistributeShell app you can set this by parameter:*--container_retry_policy*.
IMO,{code}yarn.nodemanager.recovery.enabled=true{code} and {code}ContainerRetryPolicy= NEVER_RETRY{code} is not ambivalent.
I think ContainerRetryPolicy is create to let app control which container should retry which not.
For example in ApplicationMaster assemble ContainerLaunchContext, can set this.
was (Author: feng yuan):
Hi, wy {code}ContainerRetryPolicy{code} is configuarable,for example if you are using DistributeShell app you can set this by parameter:*--container_retry_policy*.
IMO,{code}yarn.nodemanager.recovery.enabled=true{code} and {code}ContainerRetryPolicy= NEVER_RETRY{code} is not ambivalent.
I think ContainerRetryPolicy is create to let app control which container should retry which not.
> Container worker dir could not recover when NM restart
> ------------------------------------------------------
>
> Key: YARN-6630
> URL: https://issues.apache.org/jira/browse/YARN-6630
> Project: Hadoop YARN
> Issue Type: Bug
> Reporter: Yang Wang
> Attachments: YARN-6630.001.patch
>
>
> When yarn.nodemanager.recovery.enabled is true and ContainerRetryPolicy is NEVER_RETRY, container worker dir will not be saved in NM state store.
> {code:title=ContainerLaunch.java}
> ...
> private void recordContainerWorkDir(ContainerId containerId,
> String workDir) throws IOException{
> container.setWorkDir(workDir);
> if (container.isRetryContextSet()) {
> context.getNMStateStore().storeContainerWorkDir(containerId, workDir);
> }
> }
> {code}
> Then NM restarts, container.workDir is null, and may cause other exceptions.
> {code:title=ContainerImpl.java}
> static class ResourceLocalizedWhileRunningTransition
> extends ContainerTransition {
> ...
> String linkFile = new Path(container.workDir, link).toString();
> ...
> {code}
> {code}
> java.lang.IllegalArgumentException: Can not create a Path from a null string
> at org.apache.hadoop.fs.Path.checkPathArg(Path.java:159)
> at org.apache.hadoop.fs.Path.<init>(Path.java:175)
> at org.apache.hadoop.fs.Path.<init>(Path.java:110)
> ... ...
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: yarn-issues-help@hadoop.apache.org