[
https://issues.apache.org/jira/browse/SOLR-7464?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14512275#comment-14512275
]
Shalin Shekhar Mangar commented on SOLR-7464:
---------------------------------------------
How do we prevent stale responses? Earlier, if a node is not connected to ZK,
it will return a 5xx error response but it looks like with this patch an old
cluster state will be returned?
> Cluster status API call should be served by the node which it was called from
> -----------------------------------------------------------------------------
>
> Key: SOLR-7464
> URL: https://issues.apache.org/jira/browse/SOLR-7464
> Project: Solr
> Issue Type: Improvement
> Reporter: Varun Thacker
> Assignee: Varun Thacker
> Priority: Minor
> Fix For: Trunk, 5.2
>
> Attachments: SOLR-7464.patch
>
>
> Currently the cluster status api call has to go to the
> OverseerCollectionProcessor to serve the request. We should directly serve
> the request from the node where the call was made to.
> That way we will end up putting lesser tasks into the overseer.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
