You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lucene.apache.org by "Shalin Shekhar Mangar (JIRA)" <ji...@apache.org> on 2018/03/02 19:16:00 UTC
[jira] [Resolved] (SOLR-12023) Autoscaling policy engine shuffles
replicas needlessly and can also suggest nonexistent replicas to be moved
[ https://issues.apache.org/jira/browse/SOLR-12023?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Shalin Shekhar Mangar resolved SOLR-12023.
------------------------------------------
Resolution: Duplicate
The test failure no longer reproduces after SOLR-12031 was fixed so I am closing this issue as a duplicate.
> Autoscaling policy engine shuffles replicas needlessly and can also suggest nonexistent replicas to be moved
> ------------------------------------------------------------------------------------------------------------
>
> Key: SOLR-12023
> URL: https://issues.apache.org/jira/browse/SOLR-12023
> Project: Solr
> Issue Type: Bug
> Security Level: Public(Default Security Level. Issues are Public)
> Components: AutoScaling, SolrCloud
> Reporter: Shalin Shekhar Mangar
> Priority: Major
> Fix For: master (8.0), 7.3
>
> Attachments: SOLR-11066-failing.patch
>
>
> A test that I wrote in SOLR-11066 found the following problem:
> Cluster: 2 nodes
> Collection: 1 shard, 3 replicas, maxShardsPerNode=5
> No autoscaling policy or preference applied
> When the trigger runs, the computed plan needlessly shuffles all three replicas and then proceeds to return suggestions with only numbers as core names. These cores do not exist. I found that these numbers are generated internally by the framework as placeholders for moved cores for further calculations. They should never ever be suggested to the users.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org