You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
[#|2024-03-04T10:54:56.452677+01:00|WARNING|GF 7.0.10|jakarta.enterprise.web.vs.server|_ThreadID=80;_ThreadName=http-listener-1(5);_LevelValue=900;|
StandardWrapperValve[com.netinium.test.rest.RESTConfiguration]: Servlet.service() for servlet com.netinium.test.rest.RESTConfiguration threw exception
java.lang.IllegalStateException: state.isBusy() : false
at com.sun.enterprise.resource.pool.ConnectionPool.resourceClosed(ConnectionPool.java:978)
at com.sun.enterprise.resource.pool.PoolManagerImpl.putbackResourceToPool(PoolManagerImpl.java:414)
at com.sun.enterprise.resource.pool.PoolManagerImpl.resourceClosed(PoolManagerImpl.java:368)
at com.sun.enterprise.resource.listener.LocalTxConnectionEventListener.connectionClosed(LocalTxConnectionEventListener.java:71)
at com.sun.gjc.spi.ManagedConnectionImpl.connectionClosed(ManagedConnectionImpl.java:735)
at com.sun.gjc.spi.base.ConnectionHolder.close(ConnectionHolder.java:193)
at com.sun.gjc.spi.jdbc40.ConnectionHolder40.close(ConnectionHolder40.java:590)
at com.mycompany.database.firmware.dbbase.FirmwareFileDbBase.selectFirmwareFileValue(FirmwareFileDbBase.java:188)
and now I also observed again as part of my investigation of Glassfish #24805
Steps to reproduce
Difficult, see #24805, suspecting resource handles to be 'compromised'.
Impact of issue
Incorrect cleanup of resources, and long stacktraces
While expecting at least to cleanup the resource, even if state is incorrect and log a warning instead of throwing an exception.
Log the item and no longer fail with an exception, but let the method finish gracefully.
Like:
if (!state.isBusy()) {
// Do not throw exception, the current transaction should not fail if the state is incorrect.
// This method is for example also called from putbackBadResourceToPool
// throw new IllegalStateException("state.isBusy() : false");
LOG.log(WARNING, "resourceClosed - Expecting 'state.isBusy() : false', but was true for handle: " + handle);
}
The text was updated successfully, but these errors were encountered:
escay
added a commit
to escay/glassfish
that referenced
this issue
Mar 6, 2024
Environment Details
Problem Description
which was also seen in my logged Payara issue:
payara/Payara#3029
and now I also observed again as part of my investigation of Glassfish #24805
Steps to reproduce
Difficult, see #24805, suspecting resource handles to be 'compromised'.
Impact of issue
Incorrect cleanup of resources, and long stacktraces
While expecting at least to cleanup the resource, even if state is incorrect and log a warning instead of throwing an exception.
I plan to add a similar fix as done in Payara in the past: payara/Payara@f1cbfc0
Like:
The text was updated successfully, but these errors were encountered: