You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-issues@hadoop.apache.org by "Zhe Zhang (JIRA)" <ji...@apache.org> on 2016/10/29 00:49:59 UTC
[jira] [Reopened] (HADOOP-10597) RPC Server signals backoff to
clients when all request queues are full
[ https://issues.apache.org/jira/browse/HADOOP-10597?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Zhe Zhang reopened HADOOP-10597:
--------------------------------
Thanks [~mingma] for the nice work!
Since the umbrella JIRA HADOOP-9640 (and the FairCallQueue feature) is available in 2.7 and 2.6, I think it makes sense to backport this to at least 2.7. Reopening to test branch-2.7 patch. Please let me know if you have any concern about adding this to branch-2.7.
> RPC Server signals backoff to clients when all request queues are full
> ----------------------------------------------------------------------
>
> Key: HADOOP-10597
> URL: https://issues.apache.org/jira/browse/HADOOP-10597
> Project: Hadoop Common
> Issue Type: Sub-task
> Reporter: Ming Ma
> Assignee: Ming Ma
> Fix For: 2.8.0, 3.0.0-alpha1
>
> Attachments: HADOOP-10597-2.patch, HADOOP-10597-3.patch, HADOOP-10597-4.patch, HADOOP-10597-5.patch, HADOOP-10597-6.patch, HADOOP-10597-branch-2.7.patch, HADOOP-10597.patch, MoreRPCClientBackoffEvaluation.pdf, RPCClientBackoffDesignAndEvaluation.pdf
>
>
> Currently if an application hits NN too hard, RPC requests be in blocking state, assuming OS connection doesn't run out. Alternatively RPC or NN can throw some well defined exception back to the client based on certain policies when it is under heavy load; client will understand such exception and do exponential back off, as another implementation of RetryInvocationHandler.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: common-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: common-issues-help@hadoop.apache.org