lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Mark Miller (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (SOLR-8368) A SolrCore needs to replay it's tlog before the leader election process.
Date Tue, 08 Dec 2015 05:08:11 GMT

    [ https://issues.apache.org/jira/browse/SOLR-8368?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15046364#comment-15046364
] 

Mark Miller commented on SOLR-8368:
-----------------------------------

bq. 1

Probably the best thing that could be done is multithreaded replay. As far as anything being
a lot better on a shared filesystem, that doesn't help me at all for general issues like this.

bq. 2

The current leader election system is a lot more involved than simply looking at what replica
has the greatest version for it’s last update.

Perhaps when the min replication param is fully first class and full featured we can considering
getting fairly wild in how we deal with this, but in the current state of things, we are careful
to minimize data loss as much as possible

By the way, this would really going to complicate / upset my shared index / tlog on a shared
filesystem solution

> A SolrCore needs to replay it's tlog before the leader election process.
> ------------------------------------------------------------------------
>
>                 Key: SOLR-8368
>                 URL: https://issues.apache.org/jira/browse/SOLR-8368
>             Project: Solr
>          Issue Type: Bug
>            Reporter: Mark Miller
>
> If we do it after like now, the correct leader may not be able to become 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


Mime
View raw message