[jira] [Commented] (IGNITE-13976) WAL disable/enable with node restarts results in mismatching state, data loss

["Ilya Kasnacheev (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/IGNITE-13976?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17262787#comment-17262787 ] 

Ilya Kasnacheev commented on IGNITE-13976:
------------------------------------------

[~av] can you please review the PR, hint at what is the root cause of the problem? The main reproducer is WalDisableTest (a main() runnable program).

I have devised a sequential test (see the PR) and fixed it with some hack, but it did not fix the root cause. I have also tried to patch in another place, but it did not fix the reproducer.

I don't understand why on a fresh node there's org.apache.ignite.internal.processors.cache.CacheGroupDescriptor#walChangeReqs if WAL change is PME and is purely sequential. I also don't understand why the WAL status is not propagated from cache information to cache group.

> WAL disable/enable with node restarts results in mismatching state, data loss
> -----------------------------------------------------------------------------
>
>                 Key: IGNITE-13976
>                 URL: https://issues.apache.org/jira/browse/IGNITE-13976
>             Project: Ignite
>          Issue Type: Bug
>          Components: cache
>    Affects Versions: 2.9.1
>            Reporter: Ilya Kasnacheev
>            Assignee: Ilya Kasnacheev
>            Priority: Major
>
> If you try to enable/disable WAL on unstable topology, you will get to state when WAL status is undefined, nodes might have different wall status and the only way to fix it is to restart the cluster, which will lead to data loss because ignite removes data if WAL is disabled on restart.
> See the reproducer in PR.



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