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 "Lei (Eddy) Xu (JIRA)" <ji...@apache.org> on 2018/04/05 19:37:00 UTC
[jira] [Updated] (HADOOP-15311) HttpServer2 needs a way to
configure the acceptor/selector count
[ https://issues.apache.org/jira/browse/HADOOP-15311?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Lei (Eddy) Xu updated HADOOP-15311:
-----------------------------------
Fix Version/s: (was: 3.0.2)
3.0.3
> HttpServer2 needs a way to configure the acceptor/selector count
> ----------------------------------------------------------------
>
> Key: HADOOP-15311
> URL: https://issues.apache.org/jira/browse/HADOOP-15311
> Project: Hadoop Common
> Issue Type: Improvement
> Components: common
> Reporter: Erik Krogen
> Assignee: Erik Krogen
> Priority: Major
> Fix For: 3.1.0, 3.0.3
>
> Attachments: HADOOP-15311.000.patch, HADOOP-15311.001.patch, HADOOP-15311.002.patch
>
>
> HttpServer2 starts up with some number of acceptors and selectors, but only allows for the automatic configuration of these based off of the number of available cores:
> {code:title=org.eclipse.jetty.server.ServerConnector}
> selectors > 0 ? selectors : Math.max(1, Math.min(4, Runtime.getRuntime().availableProcessors() / 2)))
> {code}
> {code:title=org.eclipse.jetty.server.AbstractConnector}
> if (acceptors < 0) {
> acceptors = Math.max(1, Math.min(4, cores / 8));
> }
> {code}
> A thread pool is started of size, at minimum, {{acceptors + selectors + 1}}, so in addition to allowing for a higher tuning value under heavily loaded environments, adding configurability for this enables tuning these values down in resource constrained environments such as a MiniDFSCluster.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: common-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: common-issues-help@hadoop.apache.org