[
https://issues.apache.org/jira/browse/HDFS-15757?focusedWorklogId=546216&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-546216
]
ASF GitHub Bot logged work on HDFS-15757:
-----------------------------------------
Author: ASF GitHub Bot
Created on: 02/Feb/21 19:14
Start Date: 02/Feb/21 19:14
Worklog Time Spent: 10m
Work Description: fengnanli commented on pull request #2651:
URL: https://github.com/apache/hadoop/pull/2651#issuecomment-771903717
> Thanks @fengnanli for you works here. Leave some nit comment inline.
> Sorry I do not get why the change can reduce connections here after review
the changes, is it related "Be greedy here to close as many connections as
possible in one shot"? It will be helpful if we add some javadocs explicitly.
Thanks.
Thanks for the review @Hexiaoqiao I put the reason behind this change in
the design doc in the original JIRA ticket. In short, I did synchronous
connection closing + better picking connections + greedy closing connections. I
have seen 50% reduce in number of connections and better ProxyTime. It will be
great if you can try in your setup as well.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
Issue Time Tracking
-------------------
Worklog Id: (was: 546216)
Time Spent: 2h (was: 1h 50m)
> RBF: Improving Router Connection Management
> -------------------------------------------
>
> Key: HDFS-15757
> URL: https://issues.apache.org/jira/browse/HDFS-15757
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: rbf
> Reporter: Fengnan Li
> Assignee: Fengnan Li
> Priority: Major
> Labels: pull-request-available
> Attachments: RBF_ Improving Router Connection Management_v2.pdf, RBF_
> Improving Router Connection Management_v3.pdf, RBF_ Router Connection
> Management.pdf
>
> Time Spent: 2h
> Remaining Estimate: 0h
>
> We have seen high number of connections from Router to namenodes, leaving
> namenodes unstable.
> This ticket is trying to reduce connections through some changes. Please take
> a look at the design and leave comments.
> Thanks!
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
