You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lucene.apache.org by "zengjie (JIRA)" <ji...@apache.org> on 2015/12/15 02:33:46 UTC
[jira] [Updated] (SOLR-8417) HttpSolrCall occur a Infinite loop
request process if the collection is totally not loaded
[ https://issues.apache.org/jira/browse/SOLR-8417?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
zengjie updated SOLR-8417:
--------------------------
Attachment: HttpSolrCall.patch
Add a core node check parameter pass to remoteQuery can fix this issue
> HttpSolrCall occur a Infinite loop request process if the collection is totally not loaded
> -------------------------------------------------------------------------------------------
>
> Key: SOLR-8417
> URL: https://issues.apache.org/jira/browse/SOLR-8417
> Project: Solr
> Issue Type: Bug
> Components: SolrCloud
> Affects Versions: 5.3.1
> Reporter: zengjie
> Priority: Critical
> Attachments: HttpSolrCall.patch
>
>
> If every core of a collection is not loaded by CoresLocator(for example: all core directory have no read permission).In this situation,if some request send to the collection, HttpSolrCall can't pick a available core to process the request and try to pass the request to other nodes by remoteQuery method, and every node do the same operation,this make a infinite request loop in these nodes and block a lot of http connections.
> This problem is not common problem, but if someone hit ,it's very terrible.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org