[
https://issues.apache.org/jira/browse/SOLR-7344?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14509745#comment-14509745
]
Yonik Seeley edited comment on SOLR-7344 at 4/23/15 8:45 PM:
-------------------------------------------------------------
bq. This means there is still a chance for distributed deadlocks.
Not if you keep the size of the jetty thread pool high and then reject top
level requests when we go over some high water mark.
Actually, you wouldn't even need separate thread pools... just use a countdown
latch to limit the number of threads executing at one time in any given
category (i.e. separate latch for top level request and sub-request).
I'm not against the current approach - just lamenting that it can't be done for
us auto-magically by jetty based on URL or whatever.
was (Author: [email protected]):
bq. This means there is still a chance for distributed deadlocks.
Not if you keep the size of the jetty thread pool high and then reject top
level requests when we go over some high water mark.
I'm not against the current approach - just lamenting that it can't be done for
us auto-magically by jetty based on URL or whatever.
> Use two thread pools, one for internal requests and one for external, to
> avoid distributed deadlock and decrease the number of threads that need to be
> created.
> ---------------------------------------------------------------------------------------------------------------------------------------------------------------
>
> Key: SOLR-7344
> URL: https://issues.apache.org/jira/browse/SOLR-7344
> Project: Solr
> Issue Type: Improvement
> Components: SolrCloud
> Reporter: Mark Miller
> Attachments: SOLR-7344.patch
>
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
