[jboss-jira] [JBoss JIRA] (WFWIP-318) [XP] Cluster not formed. Same test works with 7.3.0.GA image

Martin Choma (Jira) issues at jboss.org
Wed Apr 29 08:59:03 EDT 2020


     [ https://issues.redhat.com/browse/WFWIP-318?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Martin Choma updated WFWIP-318:
-------------------------------
    Steps to Reproduce:   (was: {code}mvn test -Pxp,operator -Dxtf.test_properties.path=test.properties.os-ocp42 -Dtest=OperatorKubePingTest{code})


> [XP] Cluster not formed. Same test works with 7.3.0.GA image
> ------------------------------------------------------------
>
>                 Key: WFWIP-318
>                 URL: https://issues.redhat.com/browse/WFWIP-318
>             Project: WildFly WIP
>          Issue Type: Bug
>          Components: OpenShift
>            Reporter: Martin Choma
>            Assignee: Jeff Mesnil
>            Priority: Major
>         Attachments: KUBE_PING-73-cluster-formed.log, KUBE_PING-xp-cluster-formed.log, KUBE_PING-xp-cluster-not-formed.log
>
>
> We are experiencing cluster is not formed in one of our tests [1] with XP image. You can see log of such pod in attachemnt [3]
> Same test with same configuration works Ok with 7.3.0.GA image. See log [4] 
> Test is about Operator (same version in both cases) deploying multiple instances of application running in multiple pods, such that EAP are creating cluster.
> Comparing logs I am missing {{clustering enabled}} this log message in xp case.
> {code}
> [0m11:55:44,554 INFO  [org.jgroups.protocols.kubernetes.KUBE_PING] (ServerService Thread Pool -- 62) namespace mchoma-tmp set; clustering enabled
> {code}
> Strange is we see with XP image cluster is created in non Operator scenario [5]. There are few differences in configuration:
> - JGROUPS_CLUSTER_PASSWORD: didn't helped in Operator scenario
> - port 8080 on container: can't be configured on Operator
> - port(8888, "ping"): can't be configured on Operator
> - readiness probe explicitely set: This is set on Operator as well implicitely.
> Not sure what aspect of Operator launching can prevent cluster from forming properly. 
> [1] https://gitlab.mw.lab.eng.bos.redhat.com/jbossqe-eap/openshift-eap-tests/blob/master/test-eap/src/test/java/com/redhat/xpaas/eap/operator/test/OperatorKubePingTest.java
> [2] https://issues.redhat.com/browse/EAP7-1484?focusedCommentId=14041773&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-14041773
> [3] KUBE_PING-xp-cluster-not-formed.log 
> [4] KUBE_PING-73-cluster-formed.log
> [5] KUBE_PING-xp-cluster-formed.log  
> [6] https://gitlab.mw.lab.eng.bos.redhat.com/jbossqe-eap/openshift-eap-tests/blob/master/test-eap/src/test/java/com/redhat/xpaas/eap/jgroups/NoEncryptKubePingTest.java



--
This message was sent by Atlassian Jira
(v7.13.8#713008)



More information about the jboss-jira mailing list