[jira] [Commented] (IGNITE-6290) Data loss in case of manual rebalance

["Aleksey Plekhanov (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/IGNITE-6290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17429138#comment-17429138 ] 

Aleksey Plekhanov commented on IGNITE-6290:
-------------------------------------------

[~agura], looks like no rebalance mode (\{{setRebalanceMode(NONE)}}) doesn't suppose to be used with manual rebalance. When rebalance is disabled on any PME all assigned to node partitions are force moving to the OWNING state and all not assigned to node partitions are evicted, so no further rebalance is possible. So manual rebalance is only suppose to be used with delayed rebalance (which is also deprecated functionality and planned to be removed, see IGNITE-12662).

I think it's not an issue and propose to close this ticket. What do you think?

> Data loss in case of manual rebalance
> -------------------------------------
>
>                 Key: IGNITE-6290
>                 URL: https://issues.apache.org/jira/browse/IGNITE-6290
>             Project: Ignite
>          Issue Type: Bug
>          Components: cache
>    Affects Versions: 2.1
>            Reporter: Andrey N. Gura
>            Priority: Major
>         Attachments: CacheManualRebalancingDataConsistencyTest.java
>
>
> Data can be lost due to a manual rebalancing. 
> See reproducer that always fails when topology consist of two nodes and one of them was stopped (except of case when {{NODES_CNT}} was initially 2).



--
This message was sent by Atlassian Jira
(v8.3.4#803005)