You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lucene.apache.org by "Kesharee Nandan Vishwakarma (JIRA)" <ji...@apache.org> on 2019/05/02 22:33:00 UTC
[jira] [Comment Edited] (SOLR-13410) Designated overseer not able
to become overseer quickly
[ https://issues.apache.org/jira/browse/SOLR-13410?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16832041#comment-16832041 ]
Kesharee Nandan Vishwakarma edited comment on SOLR-13410 at 5/2/19 10:32 PM:
-----------------------------------------------------------------------------
Adding a patch with mentioned test case.
In case of more than one designated overseer nodes, there will be at least one transition overseer node before a designated node assumes overseer role. This is happening because OverseerNodePrioritizer is triggered once we have a new overseer node upon stopping a designated overseer leader node.
was (Author: keshareenv):
Adding a patch with mentioned test case.
In case of more than one designated overseer nodes, there will be at least one transition overseer node before a designated node assumes overseer role. This is happening because OverseerNodePrioritizer is triggered once we have a new overseer node upon stopping a designated overseer leader node.
[~ichattopadhyaya] Is there a way we can avoid as well?
> Designated overseer not able to become overseer quickly
> --------------------------------------------------------
>
> Key: SOLR-13410
> URL: https://issues.apache.org/jira/browse/SOLR-13410
> Project: Solr
> Issue Type: Bug
> Security Level: Public(Default Security Level. Issues are Public)
> Components: SolrCloud
> Affects Versions: master (9.0)
> Reporter: Kesharee Nandan Vishwakarma
> Assignee: Ishan Chattopadhyaya
> Priority: Major
> Attachments: SOLR-13410.patch, SOLR-13410.patch, SOLR-13410.patch, overseerElection.log
>
>
> Whenever a designated overseer node is restarted and overseer role is added back if a designated node is not overseer leader following tasks take place:
> 1. one by one nodes from electionNodes list become leader and ask designated node `to come join election at head`
> 2. current overseer node Fires Quit command and exits from Overseer Loop
> 3. Next node from `Overseer Loop` becomes leader repeats steps 1,2 until designated overseer node becomes the leader
> Problem with above flow: OverseerNodePrioritizer is not able to add `designated node` at the head of electionNodes list
> Steps to reproduce:
> # Setup solrcloud with 5 nodes, including one designated overseer
> # Restart overseer container
> Attached relevant logs
--
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