[
https://issues.apache.org/jira/browse/HDFS-7452?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14238030#comment-14238030
]
Brahma Reddy Battula commented on HDFS-7452:
--------------------------------------------
Hi [~vinayrpet]
Thanks for taking look into this issue..
{quote}
there cannot be any failover done as Client is connected to only one namenode
and tries to read the list.
{quote}
can't we skip this call to Standbynamenode..? Like Client directly connect to
Active Namenode and get the list..what do you say..?
> Can we skip getCorruptFiles() call for standby NameNode..?
> ----------------------------------------------------------
>
> Key: HDFS-7452
> URL: https://issues.apache.org/jira/browse/HDFS-7452
> Project: Hadoop HDFS
> Issue Type: Bug
> Reporter: Brahma Reddy Battula
> Assignee: Brahma Reddy Battula
> Priority: Trivial
>
> Seen following warns logs from StandBy Namenode logs ..
> {noformat}
> 2014-11-27 17:50:32,497 | WARN | 512264920@qtp-429668078-606 | Get corrupt
> file blocks returned error: Operation category READ is not supported in state
> standby |
> org.apache.hadoop.hdfs.server.namenode.FSNamesystem.getCorruptFiles(FSNamesystem.java:6916)
> 2014-11-27 17:50:42,557 | WARN | 512264920@qtp-429668078-606 | Get corrupt
> file blocks returned error: Operation category READ is not supported in state
> standby |
> org.apache.hadoop.hdfs.server.namenode.FSNamesystem.getCorruptFiles(FSNamesystem.java:6916)
> 2014-11-27 17:50:52,617 | WARN | 512264920@qtp-429668078-606 | Get corrupt
> file blocks returned error: Operation category READ is not supported in state
> standby |
> org.apache.hadoop.hdfs.server.namenode.FSNamesystem.getCorruptFiles(FSNamesystem.java:6916)
> 2014-11-27 17:51:00,058 | WARN | 512264920@qtp-429668078-606 | Get corrupt
> file blocks returned error: Operation category READ is not supported in state
> standby |
> org.apache.hadoop.hdfs.server.namenode.FSNamesystem.getCorruptFiles(FSNamesystem.java:6916)
> 2014-11-27 17:51:00,117 | WARN | 512264920@qtp-429668078-606 | Get corrupt
> file blocks returned error: Operation category READ is not supported in state
> standby |
> org.apache.hadoop.hdfs.server.namenode.FSNamesystem.getCorruptFiles(FSNamesystem.java:6916)
> 2014-11-27 17:51:02,678 | WARN | 512264920@qtp-429668078-606 | Get corrupt
> file blocks returned error: Operation category READ is not supported in state
> standby |
> org.apache.hadoop.hdfs.server.namenode.FSNamesystem.getCorruptFiles(FSNamesystem.java:6916)
> 2014-11-27 17:51:12,738 | WARN | 512264920@qtp-429668078-606 | Get corrupt
> file blocks returned error: Operation category READ is not supported in state
> standby |
> org.apache.hadoop.hdfs.server.namenode.FSNamesystem.getCorruptFiles(FSNamesystem.java:6916)
> 2014-11-27 17:51:22,798 | WARN | 512264920@qtp-429668078-606 | Get corrupt
> file blocks returned error: Operation category READ is not supported in state
> standby |
> org.apache.hadoop.hdfs.server.namenode.FSNamesystem.getCorruptFiles(FSNamesystem.java:6916)
> 2014-11-27 17:51:30,058 | WARN | 512264920@qtp-429668078-606 | Get corrupt
> file blocks returned error: Operation category READ is not supported in state
> standby |
> org.apache.hadoop.hdfs.server.namenode.FSNamesystem.getCorruptFiles(FSNamesystem.java:6916)
> 2014-11-27 17:51:30,119 | WARN | 512264920@qtp-429668078-606 | Get corrupt
> file blocks returned error: Operation category READ is not supported in state
> standby |
> org.apache.hadoop.hdfs.server.namenode.FSNamesystem.getCorruptFiles(FSNamesystem.java:6916)
> {noformat}
> do we need to call for SNN..? I feel, it might not be required.can we
> maintain state wide..Please let me know, If I am wrong..
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
