[ 
https://issues.apache.org/jira/browse/HDFS-2851?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13213376#comment-13213376
 ] 

Aaron T. Myers commented on HDFS-2851:
--------------------------------------

Hey Uma, I'm not sure I would exactly qualify this as a bug, since the balancer 
does indeed eventually succeed once all DNs have sent block reports, as 
indicated by the test you did where you lowered the heartbeat period.

I agree with you that the only way to fix this issue would be to cause the DNs 
to heartbeat promptly after a failover, though that's an optimization that has 
other benefits besides just making the balancer's used space determination get 
updated faster. Perhaps we could, then, change the title of this JIRA to 
something like "HA: Optimize stale block processing by triggering block reports 
immediately on failover" ?
                
> HA: After Balancer runs, usedSpace is not balancing correctly.
> --------------------------------------------------------------
>
>                 Key: HDFS-2851
>                 URL: https://issues.apache.org/jira/browse/HDFS-2851
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>          Components: balancer, data-node, ha, name-node
>    Affects Versions: HA branch (HDFS-1623)
>            Reporter: Uma Maheswara Rao G
>            Assignee: Uma Maheswara Rao G
>         Attachments: HDFS-2851-HDFS-1623-Test.patch
>
>
> After Balancer runs, usedSpace is not balancing correctly.
> {code}
> java.util.concurrent.TimeoutException: Cluster failed to reached expected 
> values of totalSpace (current: 1500, expected: 1500), or usedSpace (current: 
> 390, expected: 300), in more than 20000 msec.
>       at 
> org.apache.hadoop.hdfs.server.balancer.TestBalancer.waitForHeartBeat(TestBalancer.java:233)
>       at 
> org.apache.hadoop.hdfs.server.balancer.TestBalancerWithHANameNodes.testBalancerWithHANameNodes(TestBalancerWithHANameNodes.java:99)
> {code}

--
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

        

Reply via email to