[
https://issues.apache.org/jira/browse/HDFS-1325?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12894456#action_12894456
]
Todd Lipcon commented on HDFS-1325:
-----------------------------------
I don't quite follow. For one, the patch doesn't seem to make much sense - it
sends an OP_STATUS_TIMEOUT to the server even when the server may not have
completed sending the data, but doesn't actually close the socket on the client
side. So if the server is still trying to send data, the timeout doesn't
achieve anything, right?
I also don't quite see what you mean about causing system CPU on region server
to go up. The resource consumption is only of extra open file handles - it
shouldn't affect CPU usage at all to have idle sockets open. It's true that you
need to bump up the xceiver count and ulimit on DNs for HBase, but once you've
done that it doesn't cause big issues in practice.
There are a number of other JIRAs already open to work on the general issue of
socket efficiency - eg HDFS-918, HDFS-941, etc.
> DFSClient(DFSInputStream) release the persistent connection with datanode
> when no data have been read for a long time
> ---------------------------------------------------------------------------------------------------------------------
>
> Key: HDFS-1325
> URL: https://issues.apache.org/jira/browse/HDFS-1325
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: hdfs client
> Reporter: jinglong.liujl
> Fix For: 0.20.3
>
> Attachments: dfsclient.patch
>
>
> When you use Hbase over hadoop. We found during scanning over a large table (
> which has many regions and each region has many store files), there're too
> many connections has been kept between regionserver (act as DFSClient) and
> datanode. Even if the store file has been complete to scanning, the
> connections can not be closed.
> In our cluster, too many extra connections cause too many system resource has
> been wasted, which cause system cpu on region server reach to a high level,
> then bring this region server down.
> After investigating, we found the number of active connection is very small,
> and the most connection is idle. We add a timeout checker thread into
> DFSClient, to close this connection.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
