You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ignite.apache.org by "Aleksandr Polovtcev (Jira)" <ji...@apache.org> on 2023/08/21 11:02:00 UTC

[jira] [Updated] (IGNITE-20259) ClusterTimeTest#testSchedulerProlongation is flaky on TC

     [ https://issues.apache.org/jira/browse/IGNITE-20259?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Aleksandr Polovtcev updated IGNITE-20259:
-----------------------------------------
    Description: 
{{org.apache.ignite.internal.metastorage.server.time.ClusterTimeTest.testSchedulerProlongation(MetaStorageConfiguration)}} fails sometimes on TC with the following error:

{noformat}
org.mockito.exceptions.verification.MoreThanAllowedActualInvocations:
Wanted at most 0 times but was 1
  at app//org.apache.ignite.internal.metastorage.server.time.ClusterTimeTest.testSchedulerProlongation(ClusterTimeTest.java:124)
{noformat}

I'm pretty sure that the problem is that the test relies on some tight timings (e.g. that no idle safe time has been propagated in 20 ms), which may be offset some events, such as GC pauses. This test should be refactored to avoid such timings.

  was:
{{org.apache.ignite.internal.metastorage.server.time.ClusterTimeTest.testSchedulerProlongation(MetaStorageConfiguration)}} fails sometimes on TC with the following error:

{noformat}
org.mockito.exceptions.verification.MoreThanAllowedActualInvocations:
Wanted at most 0 times but was 1
  at app//org.apache.ignite.internal.metastorage.server.time.ClusterTimeTest.testSchedulerProlongation(ClusterTimeTest.java:124)
{noformat}

I'm pretty sure that the problem is that the test relies on some tight timings (e.g. that not idle safe time has been propagated in 20 ms), which may be offset some events, such as GC pauses. This test should be refactored to avoid such timings.


> ClusterTimeTest#testSchedulerProlongation is flaky on TC
> --------------------------------------------------------
>
>                 Key: IGNITE-20259
>                 URL: https://issues.apache.org/jira/browse/IGNITE-20259
>             Project: Ignite
>          Issue Type: Bug
>            Reporter: Aleksandr Polovtcev
>            Assignee: Aleksandr Polovtcev
>            Priority: Major
>              Labels: ignite-3
>
> {{org.apache.ignite.internal.metastorage.server.time.ClusterTimeTest.testSchedulerProlongation(MetaStorageConfiguration)}} fails sometimes on TC with the following error:
> {noformat}
> org.mockito.exceptions.verification.MoreThanAllowedActualInvocations:
> Wanted at most 0 times but was 1
>   at app//org.apache.ignite.internal.metastorage.server.time.ClusterTimeTest.testSchedulerProlongation(ClusterTimeTest.java:124)
> {noformat}
> I'm pretty sure that the problem is that the test relies on some tight timings (e.g. that no idle safe time has been propagated in 20 ms), which may be offset some events, such as GC pauses. This test should be refactored to avoid such timings.



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