[
https://issues.apache.org/jira/browse/HDFS-2290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13104806#comment-13104806
]
Benoy Antony commented on HDFS-2290:
------------------------------------
Once the above change is made, the NN will replicate the block and if the block
is replicated, then DN will get the command to delete the corrupt replica. The
Deletion by DN will still fail with the exception specified in the previous
comments. This is because the generationstamp of the block is 0 on datanode and
will be different from the generationstamp send by the NN.
One way to fix will be to NOT compare the generationstamp since the block is
corrupt anyway.
> Block with corrupt replica is not getting replicated
> ----------------------------------------------------
>
> Key: HDFS-2290
> URL: https://issues.apache.org/jira/browse/HDFS-2290
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: name-node
> Affects Versions: 0.22.0
> Reporter: Konstantin Shvachko
> Priority: Blocker
> Fix For: 0.22.0
>
>
> A block has one replica marked as corrupt and two good ones. countNodes()
> correctly detects that there are only 2 live replicas, and fsck reports the
> block as under-replicated. But ReplicationMonitor never schedules replication
> of good replicas.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
