You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ignite.apache.org by "Raymond Wilson (Jira)" <ji...@apache.org> on 2023/12/30 23:27:00 UTC

[jira] [Commented] (IGNITE-21170) Partition divergence on instable topology in replicated cahce

    [ https://issues.apache.org/jira/browse/IGNITE-21170?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17801402#comment-17801402 ] 

Raymond Wilson commented on IGNITE-21170:
-----------------------------------------

Is it possible this issue is triggered by any topology change on the grid, not just changes to the nodes which are responsible for the replicated cache?

> Partition divergence on instable topology in replicated cahce
> -------------------------------------------------------------
>
>                 Key: IGNITE-21170
>                 URL: https://issues.apache.org/jira/browse/IGNITE-21170
>             Project: Ignite
>          Issue Type: Bug
>            Reporter: Vladislav Pyatkov
>            Assignee: Vladislav Pyatkov
>            Priority: Major
>
> The rebalancing topology for a replicated cache can be switched earlier on a particular node before all nodes rebalance. If the topology is switched on by the coordinator, it may send a full partition map with the topology, which is not reached.  This finally leads to map invoke operations on a moving partition and then divergence between replicas.



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