You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@ignite.apache.org by "Vyacheslav Koptilin (Jira)" <ji...@apache.org> on 2022/02/03 12:55:00 UTC

[jira] [Updated] (IGNITE-16465) Joining a node, that completely lost persistent state, to Ignite cluster

     [ https://issues.apache.org/jira/browse/IGNITE-16465?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Vyacheslav Koptilin updated IGNITE-16465:
-----------------------------------------
    Summary: Joining a node, that completely lost persistent state,  to Ignite cluster   (was: Joining a node, that completely lost persistent data,  to Ignite cluster )

> Joining a node, that completely lost persistent state,  to Ignite cluster 
> --------------------------------------------------------------------------
>
>                 Key: IGNITE-16465
>                 URL: https://issues.apache.org/jira/browse/IGNITE-16465
>             Project: Ignite
>          Issue Type: Improvement
>            Reporter: Vyacheslav Koptilin
>            Priority: Major
>              Labels: ignite-3
>
> It looks like, local recovery and joining the node to the cluster can lead to a problem when the persistent state is completely lost (it makes sense for the in-memory cases and disk corruption) This node can participate in a leader election twice in the same term, and potentially, it can vote for different leaders.
> Need to propose a solution in order to avoid this issue.



--
This message was sent by Atlassian Jira
(v8.20.1#820001)