When shutting down Tomcat 7 it complaints that BoneCp isn't unregistering its self correctly. See below for the logging.
How Can I make sure all shutdown/unregister correctly?
In the logging below you see that ActiveMQ is shutting down correctly, but then the Spring isn't able to shutdown all correctly.
I am using Spring 3.1 and BoneCp 0.7.1.RELEASE with Postgresql.
The Spring part of bonecp contains a destroy hook:
- Code: Select all
<bean id="mainDataSource" class="com.jolbox.bonecp.BoneCPDataSource" destroy-method="close">
Logging
---
- Code: Select all
13:40:58.474 [Thread-15] INFO o.a.activemq.broker.BrokerService - ActiveMQ JMS Message Broker (localhost, ID:CyberEd-62087-1331296836374-0:1) stopped
13:40:58.498 [Thread-15] INFO o.s.s.c.ThreadPoolTaskScheduler - Shutting down ExecutorService 'taskScheduler'
13:40:58.891 [Thread-15] INFO com.jolbox.bonecp.BoneCP - Shutting down connection pool...
13:40:58.959 [Thread-15] INFO com.jolbox.bonecp.BoneCP - Connection pool has been shutdown.
9-mrt-2012 13:40:58 org.apache.catalina.loader.WebappClassLoader clearReferencesJdbc
SEVERE: The web application [/bv-web] registered the JDBC driver [org.postgresql.Driver] but failed to unregister it when the web application was stopped. To prevent a memory leak, the JDBC Driver has been forcibly unregistered.
9-mrt-2012 13:40:58 org.apache.catalina.loader.WebappClassLoader clearReferencesThreads
SEVERE: The web application [/bv-web] appears to have started a thread named [BoneCP-release-thread-helper-thread] but has failed to stop it. This is very likely to create a memory leak.
9-mrt-2012 13:40:58 org.apache.catalina.loader.WebappClassLoader clearReferencesThreads
SEVERE: The web application [/bv-web] appears to have started a thread named [BoneCP-release-thread-helper-thread] but has failed to stop it. This is very likely to create a memory leak.
9-mrt-2012 13:40:58 org.apache.catalina.loader.WebappClassLoader clearReferencesThreads
SEVERE: The web application [/bv-web] appears to have started a thread named [BoneCP-keep-alive-scheduler] but has failed to stop it. This is very likely to create a memory leak.
Thanks,
Ed