[
https://issues.apache.org/jira/browse/TRB-96?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Siegfried Goeschl resolved TRB-96.
----------------------------------
Resolution: Fixed
> Provide a more graceful shutdown of Fulcrum services
> ----------------------------------------------------
>
> Key: TRB-96
> URL: https://issues.apache.org/jira/browse/TRB-96
> Project: Turbine
> Issue Type: Bug
> Components: Fulcrum
> Reporter: Siegfried Goeschl
> Assignee: Siegfried Goeschl
>
> Having a rather complex application setup with Quartz jobs scheduled to
> invoke service
> * Application is terminating but termination takes a while (around 30 secs)
> * Quartz job triggers and checks if the service to be invoked is still
> available using ServiceManager.hasService(String). We are in the middle of a
> container disposal but hasService() happily returns true
> * Quartz job invokes the service but during the disposal we get a lot of
> exceptions
> A better approach (but not bullet-proof) is
> * returning "false" from ServiceManager.hasService(String) when the container
> is currently disposing. That would give my Quartz job the possibility to skip
> the job execution
> * service lookup still needs to work in my case
> * make the time to start the container disposal configurable in the container
> configuration with a default value of "0"
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
