[ 
https://issues.apache.org/jira/browse/TOMEE-4655?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Markus Jung updated TOMEE-4655:
-------------------------------
    Description: 
h2. Description
When a CDI or EJB deployment fails, TomEE does not clean up the deployment ID 
it registered for that app. The stale ID stays registered, so the next app that 
reuses it fails with {{DuplicateDeploymentIdException}} before its own 
lifecycle, managed-bean, or concurrency checks even run.

This turns one bad deployment into a cascade of unrelated test failures in 
whichever app deploys next with a matching ID.

h2. Steps to reproduce / TCK reference
Excluded in {{runner-webprofile/exclusions/enterprise-beans-30.txt}} in the 
apache/tomee-tck harness repo:
* {{**/interceptor/singleton/lifecycle/descriptor/*Test.java}}
* {{**/packaging/war/mbean/interceptor/lifecycleejbcdi/*Test.java}}
* {{**/singleton/concurrency/container/annotated/*Test.java}}
* {{**/singleton/concurrency/container/inheritance/*Test.java}}

Once fixed, remove the matching lines and confirm a deployment that follows a 
failed one with the same ID no longer fails with 
{{DuplicateDeploymentIdException}}.

  was:
h2. Summary
A failed CDI or EJB deployment does not clean up its deployment ID, so the next 
app that reuses the ID fails to deploy.

h2. Description
When a CDI or EJB deployment fails, TomEE does not clean up the deployment ID 
it registered for that app. The stale ID stays registered, so the next app that 
reuses it fails with {{DuplicateDeploymentIdException}} before its own 
lifecycle, managed-bean, or concurrency checks even run.

This turns one bad deployment into a cascade of unrelated test failures in 
whichever app deploys next with a matching ID.

h2. Steps to reproduce / TCK reference
Excluded in {{runner-webprofile/exclusions/enterprise-beans-30.txt}} in the 
apache/tomee-tck harness repo:
* {{**/interceptor/singleton/lifecycle/descriptor/*Test.java}}
* {{**/packaging/war/mbean/interceptor/lifecycleejbcdi/*Test.java}}
* {{**/singleton/concurrency/container/annotated/*Test.java}}
* {{**/singleton/concurrency/container/inheritance/*Test.java}}

Once fixed, remove the matching lines and confirm a deployment that follows a 
failed one with the same ID no longer fails with 
{{DuplicateDeploymentIdException}}.


> A failed CDI/EJB deployment leaks its deployment ID into later apps
> -------------------------------------------------------------------
>
>                 Key: TOMEE-4655
>                 URL: https://issues.apache.org/jira/browse/TOMEE-4655
>             Project: TomEE
>          Issue Type: Bug
>            Reporter: Markus Jung
>            Priority: Major
>
> h2. Description
> When a CDI or EJB deployment fails, TomEE does not clean up the deployment ID 
> it registered for that app. The stale ID stays registered, so the next app 
> that reuses it fails with {{DuplicateDeploymentIdException}} before its own 
> lifecycle, managed-bean, or concurrency checks even run.
> This turns one bad deployment into a cascade of unrelated test failures in 
> whichever app deploys next with a matching ID.
> h2. Steps to reproduce / TCK reference
> Excluded in {{runner-webprofile/exclusions/enterprise-beans-30.txt}} in the 
> apache/tomee-tck harness repo:
> * {{**/interceptor/singleton/lifecycle/descriptor/*Test.java}}
> * {{**/packaging/war/mbean/interceptor/lifecycleejbcdi/*Test.java}}
> * {{**/singleton/concurrency/container/annotated/*Test.java}}
> * {{**/singleton/concurrency/container/inheritance/*Test.java}}
> Once fixed, remove the matching lines and confirm a deployment that follows a 
> failed one with the same ID no longer fails with 
> {{DuplicateDeploymentIdException}}.



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

Reply via email to