hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Uma Maheswara Rao G (Commented) (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-1241) Possible deadlock between LeaseManager and FSDirectory
Date Thu, 15 Dec 2011 15:08:30 GMT

    [ https://issues.apache.org/jira/browse/HDFS-1241?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13170266#comment-13170266
] 

Uma Maheswara Rao G commented on HDFS-1241:
-------------------------------------------

Above comments were with respective to NameNode.
We have BackupNode based HA implementation internally. Also we have the implementation of
sending the block reports from DNs to BackupNode. Here the possibility is that, when blockreceived
command processing at Backup node side, it can invoke lease.findPath() api and at the same
time if we receive the new journal call from active node to backup node, it can go into this
situation.
                
> Possible deadlock between LeaseManager and FSDirectory
> ------------------------------------------------------
>
>                 Key: HDFS-1241
>                 URL: https://issues.apache.org/jira/browse/HDFS-1241
>             Project: Hadoop HDFS
>          Issue Type: Bug
>          Components: name-node
>    Affects Versions: 0.20.2, 0.21.0
>            Reporter: Todd Lipcon
>         Attachments: leasemanager.png
>
>
> LeaseManager.findPath() locks LeaseManager, then FSDirectory by caling getFileINode.
FSDirectory.unprotectedDelete locks itself and then calls LeaseManager.removeLeaseWithPrefixPath.
This cycle could deadlock.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message