[ https://issues.apache.org/jira/browse/HDFS-2290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13091657#comment-13091657 ]
Konstantin Shvachko commented on HDFS-2290: ------------------------------------------- h3. Steps to reproduce: - Start a cluster with 3 DataNodes. I run it on one box, and I accelerate block reports, so that each DN sends them every 3 minutes. - Create several files. - Open data-node storage directory {{data/current/finalised}}, which holds block files and remove file blk_*.meta for one of the blocks. - The block will be reported to the NameNode as corrupt. - Run fsck or -metasave to see the block is under-replicated. And this never changes. > 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 > 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