[
https://issues.apache.org/jira/browse/SOLR-5439?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13824524#comment-13824524
]
Mark Miller commented on SOLR-5439:
-----------------------------------
I think I've tracked this down - it's the low socket read timeouts we are using
due to freebsd blackhole - the cmd is successful, but it timeouts before the
client finds out - so it tries the next server in it's list. Raising the test
timeout from 30 sec to 60 sec made the issue go away.
The same thing is probably happening in the jenkins fails where a single
collection create call results in two collection create calls.
> Using CloudSolrServer to delete a collection results in two attempts to
> remove the collection, the second of which fails because the collection is
> already gone.
> ----------------------------------------------------------------------------------------------------------------------------------------------------------------
>
> Key: SOLR-5439
> URL: https://issues.apache.org/jira/browse/SOLR-5439
> Project: Solr
> Issue Type: Bug
> Components: SolrCloud
> Reporter: Mark Miller
> Assignee: Mark Miller
> Fix For: 5.0, 4.7
>
>
> A non CloudSolrServer client appears to work as expected.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
