You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ignite.apache.org by " Kirill Sizov (Jira)" <ji...@apache.org> on 2024/01/11 10:20:00 UTC

[jira] [Commented] (IGNITE-21070) Ensure that data node's primary replica expiration properly handled

    [ https://issues.apache.org/jira/browse/IGNITE-21070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17805489#comment-17805489 ] 

 Kirill Sizov commented on IGNITE-21070:
----------------------------------------

Replication finishing. (Inflights == 0) case will be covered in IGNITE-21071

> Ensure that data node's primary replica expiration properly handled
> -------------------------------------------------------------------
>
>                 Key: IGNITE-21070
>                 URL: https://issues.apache.org/jira/browse/IGNITE-21070
>             Project: Ignite
>          Issue Type: New Feature
>            Reporter: Alexander Lapin
>            Assignee:  Kirill Sizov
>            Priority: Major
>              Labels: ignite-3
>          Time Spent: 10m
>  Remaining Estimate: 0h
>
> h3. Motivation
> Corresponding primary replica expiration logic should already be implemented, mainly within coordinator recovery thus within a given ticket it's only required to extend test coverage. Primary replica expiration handling logic differs whether expiration itself occurred before or after following flow splitters:
>  * Replication finishing. (Inflights == 0)
>  * Commit timestamp evolution. // Not sure actually, maybe there's no difference between replication finishing and commit timestamp evolution splitters.
>  * Finish request handling.
>  * Cleanup request handling.
>  * WriteIntent switch request handling.
> Specific test scenarious will be specified during ticket implementation.



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