You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@lucene.apache.org by "Mark Miller (JIRA)" <ji...@apache.org> on 2015/09/23 16:06:04 UTC

[jira] [Commented] (SOLR-8087) Look into defensive check in publish that will not let a replica in LIR publish ACTIVE.

    [ https://issues.apache.org/jira/browse/SOLR-8087?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14904553#comment-14904553 ] 

Mark Miller commented on SOLR-8087:
-----------------------------------

It may be that this is just a fail safe and that it was doing it's job due to SOLR-8069?

My test that tickles this in SOLR-8075 was pre SOLR-8069 and simulates and an issue you should not run into with it (hopefully). But I was a little concerned we ended up hitting this failsafe rather than something that would allow another replica attempt leadership. 

> Look into defensive check in publish that will not let a replica in LIR publish ACTIVE.
> ---------------------------------------------------------------------------------------
>
>                 Key: SOLR-8087
>                 URL: https://issues.apache.org/jira/browse/SOLR-8087
>             Project: Solr
>          Issue Type: Bug
>            Reporter: Mark Miller
>
> What I am worried about here is that if you hit this situation, how is the election canceled? It seems like perhaps the leader can't publish ACTIVE and then the shard is locked even if another replica could be leader?



--
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