[
https://issues.apache.org/jira/browse/HDFS-5346?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13792788#comment-13792788
]
Kihwal Lee commented on HDFS-5346:
----------------------------------
I propose removing initializeReplQueues() from checkMode(). If IBR is handled
in safe mode. Also in checkMode(), call to needEnter() causes
DatanodeManager#getNumLiveDataNodes() to be called. So it is being called for
every block being added. We can modify the condition so that it is skipped if
the dn threshold is 0. DatanodeManager#getNumLiveDataNodes() walks the TreeMap
and recounts the live nodes.
> Replication queues should not be initialized in the middle of IBR processing.
> -----------------------------------------------------------------------------
>
> Key: HDFS-5346
> URL: https://issues.apache.org/jira/browse/HDFS-5346
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: namenode, performance
> Affects Versions: 0.23.9, 2.3.0
> Reporter: Kihwal Lee
> Fix For: 2.3.0, 0.23.10
>
>
> When initial block reports are being processed, checkMode() is called from
> incrementSafeBlockCount(). This causes the replication queues to be
> initialized in the middle of processing a block report in the IBR processing
> mode. If there are many block reports waiting to be processed,
> SafeModeMonitor won't be able to make name node leave the safe mode soon. It
> appears that the block report processing speed degrades considerably during
> this time.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
