[
https://issues.apache.org/jira/browse/SOLR-7280?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15380062#comment-15380062
]
Shalin Shekhar Mangar commented on SOLR-7280:
---------------------------------------------
What was the motivation behind changing the sorting logic? Can you please
explain what use-cases does the new logic cover better?
bq. The shards with least no:of replicas in down nodes and there is at least
one live node waiting for replicas of this shard. If these nodes are in down
nodes, it is no use bringing up a replica because, until those down nodes come
up, that shard cannot be up
This is flawed because with these rules, a shard which has exactly 1 replica in
total and that too on the current node will always be chosen last.
> Load cores in sorted order and tweak coreLoadThread counts to improve cluster
> stability on restarts
> ---------------------------------------------------------------------------------------------------
>
> Key: SOLR-7280
> URL: https://issues.apache.org/jira/browse/SOLR-7280
> Project: Solr
> Issue Type: Bug
> Components: SolrCloud
> Reporter: Shalin Shekhar Mangar
> Assignee: Noble Paul
> Fix For: 5.2, 6.0
>
> Attachments: SOLR-7280.patch, SOLR-7280.patch
>
>
> In SOLR-7191, Damien mentioned that by loading solr cores in a sorted order
> and tweaking some of the coreLoadThread counts, he was able to improve the
> stability of a cluster with thousands of collections. We should explore some
> of these changes and fold them into Solr.
--
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
