You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lucene.apache.org by "André Cruz (JIRA)" <ji...@apache.org> on 2014/01/15 17:11:21 UTC
[jira] [Issue Comment Deleted] (SOLR-3606) Set the default timeout
of HttpClient to a nonzero value
[ https://issues.apache.org/jira/browse/SOLR-3606?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
André Cruz updated SOLR-3606:
-----------------------------
Comment: was deleted
(was: So, "10K threads will be enough for everyone"? If we cannot know the upper bound of this timeout value at least it should be configurable so that users can insert the correct value.)
> Set the default timeout of HttpClient to a nonzero value
> --------------------------------------------------------
>
> Key: SOLR-3606
> URL: https://issues.apache.org/jira/browse/SOLR-3606
> Project: Solr
> Issue Type: Bug
> Components: SolrCloud
> Affects Versions: 5.0
> Reporter: jiangwen wei
> Attachments: SOLR-3606.patch
>
>
> The default timeout of HttpClient in HttpShardHandlerFactory and SolrCmdDistributor is set to zero.
> Zero timeout means infinite timeout, which may cause infinite waiting.
> Considering the following case which is observed in our solr cluster:
> There are two servers A and B in solr cluster with two shards.
> Server A receive a search request from client and send a sub request to server B.
> Server B also receive a search request from client and send a sub request to server A.
> the two requests cannot be completed forever, if the threads of jetty server in server A and server B exhausted.
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org