ode-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Tammo van Lessen (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (ODE-897) Fix the undeployment logic
Date Fri, 22 Jul 2011 11:40:57 GMT

    [ https://issues.apache.org/jira/browse/ODE-897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13069511#comment-13069511
] 

Tammo van Lessen commented on ODE-897:
--------------------------------------

But isn't that the expected behavior? If a process is retired it must still be able to finish
instances that have already been started. So, if some of these instances are still waiting
for messages at that endpoint, it must remain activated. Surely it can be removed once all
instances have been removed, but that is then not part of the undeployment logic.

> Fix the undeployment logic
> --------------------------
>
>                 Key: ODE-897
>                 URL: https://issues.apache.org/jira/browse/ODE-897
>             Project: ODE
>          Issue Type: Bug
>          Components: Deployment
>    Affects Versions: 1.3.5
>         Environment: Ubuntu 10.10
> java 6
>            Reporter: Waruna Ranasinghe
>            Assignee: Jeff Yu
>
> When a process is undeployed, in BpelProcess#deactivate() method, it checks whether the
process is ACTIVE and call deactivateMyRoleEndpoint(endpoint).
> But, what if the process is retired and this process is the only process of its type,
then the deactivateMyRoleEndpoint(endpoint) will not be called and the respective axis service
will not be removed.

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message