[jboss-jira] [JBoss JIRA] (WFLY-8540) Former singleton provider node leaving cluster after re-election causes another re-election
[
https://issues.jboss.org/browse/WFLY-8540?page=com.atlassian.jira.plugin....
]
Paul Ferraro moved JBEAP-10216 to WFLY-8540:
--------------------------------------------
Project: WildFly (was: JBoss Enterprise Application Platform)
Key: WFLY-8540 (was: JBEAP-10216)
Workflow: GIT Pull Request workflow (was: CDW with loose statuses v1)
Component/s: Clustering
(was: Clustering)
Affects Version/s: 11.0.0.Alpha1
(was: 7.1.0.DR15)
Former singleton provider node leaving cluster after re-election
causes another re-election
--------------------------------------------------------------------------------------------
Key: WFLY-8540
URL: https://issues.jboss.org/browse/WFLY-8540
Project: WildFly
Issue Type: Bug
Components: Clustering
Affects Versions: 11.0.0.Alpha1
Reporter: Paul Ferraro
Assignee: Paul Ferraro
Priority: Critical
With a random election policy, this situation is likely to happen:
1. After a topology change, new singleton provider is elected
2. Former singleton provider undeploys the singleton application, which causes it to
leave a cluster for a while
3. This topology change triggers another election
4. Loop continues
--
This message was sent by Atlassian JIRA
(v7.2.3#72005)