[jboss-jira] [JBoss JIRA] (AS7-6761) Deployment replace causes invalid state in DUP

[RH Bugzilla Integration (JIRA)]

    [ https://issues.jboss.org/browse/AS7-6761?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12764027#comment-12764027 ] 

RH Bugzilla Integration commented on AS7-6761:
----------------------------------------------

Stuart Douglas <sdouglas at redhat.com> made a comment on [bug 924562|https://bugzilla.redhat.com/show_bug.cgi?id=924562]

This is resolved for non-OSGi deployments. I think we should resolve this issue and create a new one for the OSGi problems, as the OSGi issue is not so much caused by dependency changes but rather by explicit start/stop operations on the bundle.
                
> Deployment replace causes invalid state in DUP
> ----------------------------------------------
>
>                 Key: AS7-6761
>                 URL: https://issues.jboss.org/browse/AS7-6761
>             Project: Application Server 7
>          Issue Type: Bug
>          Components: Server
>            Reporter: Thomas Diesler
>            Assignee: Stuart Douglas
>         Attachments: v200.jar, v201.jar, webapp-v200.war
>
>
> Webapp A depends on deployment B. Replacing B causes NPE in JPAInterceptorProcessor
> {code}
> 12:52:05,712 INFO  [org.jboss.as.server.deployment] (MSC service thread 1-1) JBAS015876: Starting deployment of "v200.jar" (runtime-name: "v200.jar")
> 12:52:05,889 INFO  [org.jboss.as.server] (XNIO-1 task-8) JBAS018559: Deployed "v200.jar" (runtime-name : "v200.jar")
> 12:52:37,207 INFO  [org.jboss.as.server.deployment] (MSC service thread 1-5) JBAS015876: Starting deployment of "webapp-v200.war" (runtime-name: "webapp-v200.war")
> 12:52:37,349 INFO  [org.jboss.web] (ServerService Thread Pool -- 52) JBAS018210: Register web context: /webapp-v200
> 12:52:37,436 INFO  [org.jboss.as.server] (XNIO-1 task-3) JBAS018559: Deployed "webapp-v200.war" (runtime-name : "webapp-v200.war")
> 12:53:20,252 INFO  [org.jboss.web] (ServerService Thread Pool -- 57) JBAS018224: Unregister web context: /webapp-v200
> 12:53:20,272 INFO  [org.jboss.as.server.deployment] (MSC service thread 1-1) JBAS015877: Stopped deployment v200.jar (runtime-name: v200.jar) in 25ms
> 12:53:20,274 INFO  [org.jboss.as.server.deployment] (MSC service thread 1-5) JBAS015876: Starting deployment of "v200.jar" (runtime-name: "v200.jar")
> 12:53:20,293 ERROR [org.jboss.msc.service.fail] (MSC service thread 1-2) MSC000001: Failed to start service jboss.deployment.unit."webapp-v200.war".FIRST_MODULE_USE: org.jboss.msc.service.StartException in service jboss.deployment.unit."webapp-v200.war".FIRST_MODULE_USE: JBAS018733: Failed to process phase FIRST_MODULE_USE of deployment "webapp-v200.war"
> 	at org.jboss.as.server.deployment.DeploymentUnitPhaseService.start(DeploymentUnitPhaseService.java:127) [jboss-as-server-8.0.0.Alpha1-SNAPSHOT.jar:8.0.0.Alpha1-SNAPSHOT]
> 	at org.jboss.msc.service.ServiceControllerImpl$StartTask.startService(ServiceControllerImpl.java:1972) [jboss-msc-1.1.1.Final.jar:1.1.1.Final]
> 	at org.jboss.msc.service.ServiceControllerImpl$StartTask.run(ServiceControllerImpl.java:1905) [jboss-msc-1.1.1.Final.jar:1.1.1.Final]
> 	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145) [rt.jar:1.7.0_13]
> 	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615) [rt.jar:1.7.0_13]
> 	at java.lang.Thread.run(Thread.java:722) [rt.jar:1.7.0_13]
> Caused by: java.lang.NullPointerException
> 	at org.jboss.as.jpa.processor.JPAInterceptorProcessor.deploy(JPAInterceptorProcessor.java:52)
> 	at org.jboss.as.server.deployment.DeploymentUnitPhaseService.start(DeploymentUnitPhaseService.java:120) [jboss-as-server-8.0.0.Alpha1-SNAPSHOT.jar:8.0.0.Alpha1-SNAPSHOT]
> 	... 5 more
> {code}
> What might be happening here is that replacing B causes ModuleB to go down. Phase FIRST_MODULE_USE of A depends on ModuleB, which causes a call to undeploy() on the DUPs for A. When ModuleB becomes available again, deploy() is called on the DUPs for A again. Possibly because of the cleanup phase, necessary state is lost and we see a NPE.
> AFAICS, this is a variation of the feature request that DUPs support multiple calls to deploy/undeploy.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira