[
https://issues.apache.org/jira/browse/HADOOP-18840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17751856#comment-17751856
]
ASF GitHub Bot commented on HADOOP-18840:
-----------------------------------------
2005hithlj commented on PR #5926:
URL: https://github.com/apache/hadoop/pull/5926#issuecomment-1668816922
> change lgtm.
>
> Also FYI, in
[HDFS-17042](https://issues.apache.org/jira/browse/HDFS-17042), we added
OverallProcessingTime for each rpc method. It measures the latency from when an
rpc request arrives at the NN to when a response is sent back. You might be
interested to check out these metrics too.
@xinglin sir.
Thanks for your review, We also need to know the exact enQeue time in order
to troubleshoot performance bottlenecks.
> Add enQueue time to RpcMetrics
> ------------------------------
>
> Key: HADOOP-18840
> URL: https://issues.apache.org/jira/browse/HADOOP-18840
> Project: Hadoop Common
> Issue Type: Improvement
> Components: rpc-server
> Reporter: Liangjun He
> Assignee: Liangjun He
> Priority: Minor
> Labels: pull-request-available
>
> Sometimes, the request time observed by the client is much longer than the
> queue + process time on the RPC server. Perhaps the RPC request 'waiting
> enQueue' took too long on the RPC server, so we should add enQueue time to
> RpcMetrics.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
