[
https://issues.apache.org/jira/browse/HDFS-9087?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14790746#comment-14790746
]
Elliott Clark commented on HDFS-9087:
-------------------------------------
On a large enough cluster anything that can thundering herd will eventually. In
this case we're seeing it on disk io and before 2.7.1 we were seeing it on
locking FsVolumeList. I suspect that we will now start to see this on block
replication load. Anything that can de-sync these across the cluster is better.
> Add some jitter to DataNode.checkDiskErrorThread
> ------------------------------------------------
>
> Key: HDFS-9087
> URL: https://issues.apache.org/jira/browse/HDFS-9087
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: datanode
> Affects Versions: 2.6.0
> Reporter: Elliott Clark
> Assignee: Elliott Clark
> Attachments: HDFS-9087-v0.patch, HDFS-9087-v1.patch
>
>
> If all datanodes are started across a cluster at the same time (or errors in
> the network cause ioexceptions) there can be storms where lots of datanodes
> check their disks at the exact same time.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
