[jira] [Commented] (SOLR-16414) Race condition in PRS state updates

["Ishan Chattopadhyaya (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SOLR-16414?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17606697#comment-17606697 ] 

Ishan Chattopadhyaya commented on SOLR-16414:
---------------------------------------------

I benchmarked the collection CREATE operations using solr-bench after this change.
It took 297 seconds to create 1000 PRS collections, vs 307 seconds to create 1000 non-PRS collections.

The suite is here: https://github.com/fullstorydev/solr-bench/blob/master/cluster-test.json (just change the correct commit to point to this PR branch, and change "perReplicaState" variable to true/false for the comparison).

> Race condition in PRS state updates
> -----------------------------------
>
>                 Key: SOLR-16414
>                 URL: https://issues.apache.org/jira/browse/SOLR-16414
>             Project: Solr
>          Issue Type: Bug
>      Security Level: Public(Default Security Level. Issues are Public) 
>            Reporter: Noble Paul
>            Priority: Major
>          Time Spent: 20m
>  Remaining Estimate: 0h
>
> For PRS collections the individual states are potentially updated from individual nodes and sometimes from overseer too. it's possible that
>  
>  # OP1 is sent to overseer at T1
>  # OP2 is executed in the node itself at T2
>  
> Because we cannot guarantee that the OP1 sent to overseer may execute before OP2 tyhe final state will be the result of OP1 which is incorrect and can lead to errors .
> The solution is to never do any PRS writes from overseer. 



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@solr.apache.org
For additional commands, e-mail: issues-help@solr.apache.org