[jira] [Commented] (IGNITE-25506) Investigate flaky ItTxDistributedTestThreeNodesThreeReplicasCollocated#testBalance

["Kirill Tkalenko (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/IGNITE-25506?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17955148#comment-17955148 ] 

Kirill Tkalenko commented on IGNITE-25506:
------------------------------------------

Added logging to [ignite-25506-debug|https://github.com/gridgain/apache-ignite-3/tree/ignite-25506-debug], and found out that the commit of the intent to write can happen in *org.apache.ignite.internal.table.distributed.StorageUpdateHandler#handleUpdate* or *org.apache.ignite.internal.table.distributed.StorageUpdateHandler#handleUpdateAll*, so in *org.apache.ignite.internal.table.distributed.StorageUpdateHandler#switchWriteIntents* it was possible to get a mismatch between the transaction that created it and the one that commits it, and this will create big problems.

> Investigate flaky ItTxDistributedTestThreeNodesThreeReplicasCollocated#testBalance
> ----------------------------------------------------------------------------------
>
>                 Key: IGNITE-25506
>                 URL: https://issues.apache.org/jira/browse/IGNITE-25506
>             Project: Ignite
>          Issue Type: Bug
>            Reporter: Kirill Tkalenko
>            Assignee: Kirill Tkalenko
>            Priority: Major
>              Labels: ignite-3
>          Time Spent: 0.5h
>  Remaining Estimate: 0h
>
> We need to investigate the reason for the fall of test *org.apache.ignite.distributed.ItTxDistributedTestThreeNodesThreeReplicasCollocated#testBalance*.
> Link with first fall, it happened after IGNITE-25477, we need to find out the reason.



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