[jira] [Commented] (IGNITE-6792) SQL fields query for huge replicated cache gives not correct count if rebalancing is in progress

["Dmitriy Pavlov (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/IGNITE-6792?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16225073#comment-16225073 ] 

Dmitriy Pavlov commented on IGNITE-6792:
----------------------------------------

I've attached reproducer and added it to branch https://github.com/apache/ignite/compare/master...gridgain:ignite-6792-test?expand=1

> SQL fields query for huge replicated cache gives not correct count if rebalancing is in progress
> ------------------------------------------------------------------------------------------------
>
>                 Key: IGNITE-6792
>                 URL: https://issues.apache.org/jira/browse/IGNITE-6792
>             Project: Ignite
>          Issue Type: Bug
>      Security Level: Public(Viewable by anyone) 
>          Components: sql
>    Affects Versions: 2.3
>            Reporter: Dmitriy Pavlov
>         Attachments: GridReplicatedCacheRebalancingWithPdsEnabled.java
>
>
> According to recent user-list question http://apache-ignite-users.70518.x6.nabble.com/Ignite-2-3-replicated-cache-lost-data-after-restart-cluster-nodes-with-persistence-enabled-td17612.html 
> reproducing test was created.
> This test creates one grid and loads 1 000 000 records to replicated cache. Persistent store is enabled.
> Then second grid is connected and rebalancing is started.
> We don't wait to rebalance is completely finished and run 
> select count(*) from Organization
> and scan query.
> Results returned by count in second node is less than initially loaded data amount
> At grid 0: Cache .size(): [512149], sql fields query count is [1000000],  sql scan query count is [1000000]
> At grid 1: Cache .size(): [512149], sql fields query count is [44218], sql scan query count is [44219]



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)