[
https://issues.apache.org/jira/browse/YARN-9157?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16732126#comment-16732126
]
Weiwei Yang edited comment on YARN-9157 at 1/2/19 3:01 PM:
-----------------------------------------------------------
Hi [~jj336013]
If I remember correctly, DeletionService should have done persistence for its
tasks in NM state store. Is that only used during NM recovery (restart)? If
that's the case, we can certainly improve this to something like you mentioned.
Are you going to work on this? I can help to review.
Thanks
was (Author: cheersyang):
Hi [~jj336013]
If I remember correctly, DeletionService should have done persistence for its
tasks in NM state store. Is that only used during NM recovery (restart)? If
that's the case, we can certainly improve this to something like you mentioned.
Are you going to work on this? I can help to review.
Thanks
> Failed deletion dirs in yarn.nodemanager.local-dirs causes accumulation
> lots of files under the path yarn.nodemanager.local-dirs and causes
> operation systerm's Inode to to be depleted
> ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
>
> Key: YARN-9157
> URL: https://issues.apache.org/jira/browse/YARN-9157
> Project: Hadoop YARN
> Issue Type: Bug
> Components: applications/distributed-shell
> Affects Versions: 2.7.2, 2.8.3, 2.7.5, 3.0.1, 3.1.1
> Reporter: yimeng
> Priority: Major
> Attachments: image-2018-12-25-10-03-51-070.png
>
>
> the Yarn task Excute failed , because excessive number of files under the
> path yarn.nodemanager.local-dirs causes Inode to run out and calculates task
> failure
> check the NM Logs , found that many localized dirs delete failed because of
> user not found in security Systerm.
> _2018-12-21 06:06:40,723 | INFO | AsyncDispatcher event handler | Cache Size
> Before Clean: 240859897, Total Deleted: 85003, Public Deleted: 0, Private
> Deleted: 85003 | ResourceLocalizationService.java:522_
> _2018-12-21 06:06:40,744 | ERROR | DeletionService #1 | DeleteAsUser for
> /srv/BigData/hadoop/data1/nm/localdir/usercache/odaeuser/filecache/48339
> returned with exit code: 255 | LinuxContainerExecutor.java:565_
> _ExitCodeException exitCode=255:_
> _at org.apache.hadoop.util.Shell.runCommand(Shell.java:664)_
> _at org.apache.hadoop.util.Shell.run(Shell.java:553)_
> _at
> org.apache.hadoop.util.Shell$ShellCommandExecutor.execute(Shell.java:866)_
> _at
> org.apache.hadoop.yarn.server.nodemanager.LinuxContainerExecutor.deleteAsUser(LinuxContainerExecutor.java:559)_
> _at
> org.apache.hadoop.yarn.server.nodemanager.DeletionService$FileDeletionTask.run(DeletionService.java:276)_
> _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$Worker.run(ThreadPoolExecutor.java:624)_
> _at java.lang.Thread.run(Thread.java:748)_
> _2018-12-21 06:06:40,744 | ERROR | DeletionService #1 | Output from
> LinuxContainerExecutor's deleteAsUser follows: |
> LinuxContainerExecutor.java:567_
> _2018-12-21 06:06:40,744 | INFO | DeletionService #1 | main : command
> provided 3 | ContainerExecutor.java:322_
> _2018-12-21 06:06:40,744 | INFO | DeletionService #1 | main : run as user is
> odaeuser | ContainerExecutor.java:322_
> _2018-12-21 06:06:40,744 | INFO | DeletionService #1 | main : requested yarn
> user is odaeuser | ContainerExecutor.java:322_
> _2018-12-21 06:06:40,744 | INFO | DeletionService #1 | User odaeuser not
> found | ContainerExecutor.java:322_
> _2018-12-21 06:06:40,745 | INFO | DeletionService #1 | Deleting absolute path
> : /srv/BigData/hadoop/data1/nm/localdir/usercache/odaeuser/filecache/48342 |
> LinuxContainerExecutor.java:543_
> _2018-12-21 06:06:40,749 | ERROR | DeletionService #2 | DeleteAsUser for
> /srv/BigData/hadoop/data1/nm/localdir/usercache/odaeuser/filecache/48334
> returned with exit code: 255 | LinuxContainerExecutor.java:565_
> _ExitCodeException exitCode=255:_
> _at org.apache.hadoop.util.Shell.runCommand(Shell.java:664)_
> _at org.apache.hadoop.util.Shell.run(Shell.java:553)_
> _at
> org.apache.hadoop.util.Shell$ShellCommandExecutor.execute(Shell.java:866)_
> _at
> org.apache.hadoop.yarn.server.nodemanager.LinuxContainerExecutor.deleteAsUser(LinuxContainerExecutor.java:559)_
> _at
> org.apache.hadoop.yarn.server.nodemanager.DeletionService$FileDeletionTask.run(DeletionService.java:276)_
> _at java.util.concurrent.Executors$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)_
> _2018-12-21 06:06:40,749 | ERROR | DeletionService #2 | Output from
> LinuxContainerExecutor's deleteAsUser follows: |
> LinuxContainerExecutor.java:567_
> _2018-12-21 06:06:40,749 | INFO | DeletionService #2 | main : command
> provided 3 | ContainerExecutor.java:322_
> _2018-12-21 06:06:40,749 | INFO | DeletionService #2 | main : run as user is
> odaeuser | ContainerExecutor.java:322_
> _2018-12-21 06:06:40,749 | INFO | DeletionService #2 | main : requested yarn
> user is odaeuser | ContainerExecutor.java:322_
> _2018-12-21 06:06:40,749 | INFO | DeletionService #2 | User odaeuser not
> found | ContainerExecutor.java:322_
>
> actually the local dir files's size is 4.4GB, not 240859897B print in the log
> !image-2018-12-25-10-03-51-070.png!
> The user not found is because of our userInfo is saved in Ldap DB , when
> Ldap Service have problem at some time , then get the user info will fail(not
> because the user is deleted).When the Ldap Server recovery at some time , the
> user info can get .
> The problem is even we can get the user info later , the dirs that deleted
> failed before will never be deleted later (it is deleted from the tracker
> list ), this cause the dirs accumulation .
> I think NM ResourceLocalizationService should determine whether the file was
> deleted successfully by Deletion Service Thread before deleting the directory
> from tracker list and levelDB,if deleted failed ,we should add back it to
> tracker list ,then delete the next dirs till the local dirs size is below
> yarn.nodemanager.localizer.cache.target-size-mb
> .
>
>
--
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
