Sathwik Bantwal Premakumar created ODE-994:
----------------------------------------------
Summary: Cleanup problem when service is destroyed with axis2 1.6
Key: ODE-994
URL: https://issues.apache.org/jira/browse/ODE-994
Project: ODE
Issue Type: Bug
Components: Axis2 Integration
Affects Versions: 1.3.5
Reporter: Sathwik Bantwal Premakumar
Assignee: Sathwik Bantwal Premakumar
Fix For: 1.3.6, 1.4
Here is the original message from Igor on the problem
(http://markmail.org/message/7i5gbdeskx3dzsyl)
After migration ODE to axis2 1.6.1 I faced with the service cleanup problem.
The problem is appeared when we redeploy process. After this redeployment
all services are cleaned up and we have only redeployed service.
I've reviewed ode-axis2 sources and found the place where this cleanup is
happened. It is destroyService method of BindingContextImpl class.
There we have axis configuration cleanup(_server._axisConfig.cleanup()) from
axis2-kernel library.
Then I analyzed changes in kernel lib of cleanup() from AxisConfiguration
class (1.3 vs 1.6) and found that in 1.6 there was added following code :
if (clusteringAgent != null) {
clusteringAgent.finalize();
}
this.policySupportedModules.clear();
this.moduleConfigmap.clear();
this.allEndpoints.clear();
this.allModules.clear();
--> this.allServices.clear();
this.outPhases.clear();
this.messageReceivers.clear();
this.targetResolvers.clear();
if (this.engagedModules != null) {
this.engagedModules.clear();
}
this.configurator = null;
Seems that we can't call _server._axisConfig.cleanup() when we use axis 1.6.
Is it correct to do cleanup of all axis config instead of cleanup only
related to destroyed service data?
If not then what should we clean for destroyed service?
Thanks,
Igor.
--
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
