[
https://issues.apache.org/jira/browse/HDFS-5927?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Brahma Reddy Battula resolved HDFS-5927.
----------------------------------------
Resolution: Not A Problem
As blocks are deleted from the datanode dir manually, all the blocks marked as
corrupted(and missed) from FSCK reported..
But Mbean attribute will be updated when replica got corrupted, this is like
replica level..Hence it's not updated the Mbean, As it is behaviour hence
closing the defect
> Corrupted blocks value is not updating in jmx(FSNamesystemMBean)
> ----------------------------------------------------------------
>
> Key: HDFS-5927
> URL: https://issues.apache.org/jira/browse/HDFS-5927
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: namenode
> Affects Versions: 2.2.0
> Reporter: Brahma Reddy Battula
> Assignee: Brahma Reddy Battula
>
> Steps -
> 1)- Write data to HDFS.
> 2)- Corrupt some blocks.
> 3)- Check whether Corrupted blocks count is updating in jmx
> Observed output - JMX is showing always count zero where fsck report had
> shown 27 Corrupted blocks are there.
> /** Returns number of blocks with corrupt replicas */
> @Metric({"CorruptBlocks", "Number of blocks with corrupt replicas"})
> public long getCorruptReplicaBlocks() {
> return blockManager.getCorruptReplicaBlocksCount();
> }
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
