[
https://issues.apache.org/jira/browse/IGNITE-10087?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16746187#comment-16746187
]
Amelchev Nikita commented on IGNITE-10087:
------------------------------------------
One of a problem was resolved in IGNITE-5115 (coordinator failing).
I have fixed the test issues:
1. Client node disconnected issue. Getting retry future can throw
{{IgniteClientDisconnectedException}}. I wrote the {{getRetryFuture}} method
and use it in the client future.
2. the {{awaitPartitionMapExchange()}} can throw "Grid is in invalid state -
node stopping). I added additional checks for servers stopped.
3. Redundant code cleanup.
I have tested the test locally and [on
TC|https://ci.ignite.apache.org/viewLog.html?buildId=2825701&tab=testsInfo&buildTypeId=IgniteTests24Java8_Spi]
more than 100 times.
The test can fail with a problem that described in IGNITE-1758 (less 5% fail
rate).
Can anyone review the test changes?
> IgniteClientReconnectMassiveShutdownTest.testMassiveServersShutdown1 is flaky
> in master
> ---------------------------------------------------------------------------------------
>
> Key: IGNITE-10087
> URL: https://issues.apache.org/jira/browse/IGNITE-10087
> Project: Ignite
> Issue Type: Bug
> Reporter: Amelchev Nikita
> Assignee: Amelchev Nikita
> Priority: Major
> Labels: MakeTeamcityGreenAgain
> Fix For: 2.8
>
>
> The IgniteClientReconnectMassiveShutdownTest.testMassiveServersShutdown1 test
> is flaky in master. Sometimes the test fails on test timeout (5 min).
> [Test
> history.|https://ci.ignite.apache.org/project.html?projectId=IgniteTests24Java8&testNameId=-6728945354254258306&branch=<default>&tab=testDetails]
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
