[
https://issues.apache.org/jira/browse/FELIX-2431?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Carsten Ziegeler resolved FELIX-2431.
-------------------------------------
Fix Version/s: eventadmin-1.4.0
Resolution: Fixed
After being hit by the same problem I fixed this in revision 989232.
The config admin update event is now handled async and does not block config
admin anymore. In addition shutting down the pools is done async as well to
make the downtime as small as possible
> EventAdmin service unregistered but not registered again on ConfigAdmin
> startup
> -------------------------------------------------------------------------------
>
> Key: FELIX-2431
> URL: https://issues.apache.org/jira/browse/FELIX-2431
> Project: Felix
> Issue Type: Bug
> Components: Event Admin
> Affects Versions: eventadmin-1.2.2
> Reporter: Thomas Diesler
> Assignee: Carsten Ziegeler
> Fix For: eventadmin-1.2.4
>
>
> In EventAdmin Configuration
> Object service = new ManagedService()
> {
> public synchronized void updated( Dictionary properties )
> throws ConfigurationException
> {
> configure( properties );
> stop();
> start();
> }
> };
> The call to stop() does not return in a timely manner (i.e. it may take >
> 40sec). This is because the asyncronous thread pool cannot be closed
> properly. This condition occurs when eventadmin is started before configadmin
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
