[
https://issues.apache.org/jira/browse/HDFS-2290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13105724#comment-13105724
]
Benoy Antony commented on HDFS-2290:
------------------------------------
The patch contains the Fixes to BlockManager as mentioned in the comments. The
bug is reproduced in a slightly different way in which the block is corrupted
by changing the contents and not by deleting the meta file. This way of
corruption allows the datanode to remove the corrupt block when asked by the
namenode.
The patch also contains a a test class which 4 unit tests testing the various
scenarios for handling corrupt replicas.
> 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
>
> Attachments: HDFS-2290_022.patch
>
>
> 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
