[jira] [Updated] (IGNITE-26638) Some tests become flaky with Not enough alive nodes to perform reset with clean up.

["Alexander Lapin (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/IGNITE-26638?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Alexander Lapin updated IGNITE-26638:
-------------------------------------
    Ignite Flags:   (was: Docs Required,Release Notes Required)

> Some tests become flaky with Not enough alive nodes to perform reset with clean up.
> -----------------------------------------------------------------------------------
>
>                 Key: IGNITE-26638
>                 URL: https://issues.apache.org/jira/browse/IGNITE-26638
>             Project: Ignite
>          Issue Type: Bug
>            Reporter: Mirza Aliev
>            Assignee: Egor Kuts
>            Priority: Major
>              Labels: MakeTeamcityGreenAgain, ignite-3
>             Fix For: 3.2
>
>          Time Spent: 4h 10m
>  Remaining Estimate: 0h
>
> h3. Motivation
> Several test in CLI tests from {{ItRestartPartitionsTest}} and REST tests from {{ItDisasterRecoveryControllerRestartPartitionsWithCleanupTest}} fail with 
> {code:java}
> io.micronaut.http.client.exceptions.HttpClientResponseException: {"title":"Internal Server Error","status":500,"code":"IGN-RECOVERY-5","detail":"Not enough alive nodes to perform reset with clean up.","traceId":"bdc79ea2-72cc-4a7b-a993-a12e4ac8047c"}
> {code}
> or similar because on a moment of invocation of restart with clean up there is no enough nodes to perform the operation. This happens because some partitions on some nodes are still in {{INITIALIZING}} state. 
> h3. 
> Possible solution is explicitly await {{HELATHY}} statues on all nodes on which we plan to perform restart with clean up right before we call CLI/REST method in every affected test



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