[
https://issues.apache.org/jira/browse/HDFS-14997?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17184920#comment-17184920
]
Xiaoqiao He commented on HDFS-14997:
------------------------------------
Thanks [~Captainhzy], In my experience, process cmds async can resolve most
cases about DataNode lost because no heartbeat for long times from NameNode
views.
In general, we have split cmds to process one by one asyc rather than process
all of them in main flow. So I believe it improved significantly.
About the lock contention, I agree that it could be blocked especially process
one very heavy cmd(maybe some corner case). Any ideas to improve it? welcome to
more discuss. Thanks again.
> BPServiceActor processes commands from NameNode asynchronously
> --------------------------------------------------------------
>
> Key: HDFS-14997
> URL: https://issues.apache.org/jira/browse/HDFS-14997
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: datanode
> Reporter: Xiaoqiao He
> Assignee: Xiaoqiao He
> Priority: Major
> Fix For: 3.3.0
>
> Attachments: HDFS-14997.001.patch, HDFS-14997.002.patch,
> HDFS-14997.003.patch, HDFS-14997.004.patch, HDFS-14997.005.patch,
> HDFS-14997.addendum.patch, image-2019-12-26-16-15-44-814.png
>
>
> There are two core functions, report(#sendHeartbeat, #blockReport,
> #cacheReport) and #processCommand in #BPServiceActor main process flow. If
> processCommand cost long time it will block send report flow. Meanwhile
> processCommand could cost long time(over 1000s the worst case I meet) when IO
> load of DataNode is very high. Since some IO operations are under
> #datasetLock, So it has to wait to acquire #datasetLock long time when
> process some of commands(such as #DNA_INVALIDATE). In such case, #heartbeat
> will not send to NameNode in-time, and trigger other disasters.
> I propose to improve #processCommand asynchronously and not block
> #BPServiceActor to send heartbeat back to NameNode when meet high IO load.
> Notes:
> 1. Lifeline could be one effective solution, however some old branches are
> not support this feature.
> 2. IO operations under #datasetLock is another issue, I think we should solve
> it at another JIRA.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
