JVM Peer Gone in WebLogic T3 connection

So I have this exception when connecting to a FileNet P8 API from my Linux Machine:

com.filenet.api.exception.EngineRuntimeException: FNRCE0040E: E_NOT_AUTHENTICATED: The user is not authenticated. Message was: java.net.ConnectException: t3://1.2.3.4:9210: Bootstrap to 1.2.3.4/1.2.3.4:9210 failed. It is likely that the remote side declared peer gone on this JVM  at com.filenet.apiimpl.core.UserPasswordToken.getSubject(UserPasswordToken.java:121)
 at com.filenet.api.util.UserContext.createSubject(UserContext.java:288)

This happens when my application is connecting to FileNet P8 Content Engine deployed into a WebLogic Cluster and not into a single node (Well it might happen with a single node…). The connection point used by the application is a single node address.

The solution for this error message?

Just add to the host files of the client machine the name and ip address of each weblogic cluster physical machine.

[K]ubuntu USB errors->Slow boot

I’ve installed Kubuntu 8.10 64 bit on my new computer. This new computer has a Asus P6T motherboard.

What started to happen was that boot became quite slow due to some USB errors… For example:

usb 5-1: device descriptor read/all, error -71 and usb 5-1: device descriptor read/all, error -110

This slowed down booting for 30s to one minute as the console was outputing these errors… After a while, boot progressed, and the KDE logon screen appeared. No issue was found accessing any connecting USB device.

First I thought of my WD Elements 400GB external drive was the reason for the issue. But fully disconnecting it didn’t do the trick. I also thought it might be an issue with the motherboard, the new Asus P6T, but it looks like it was more a Linux kernel module issue.

At least this for me worked:

rmmod uhci_hcd

rmmod  ehci_hcd

rmmod usbcore

and then change the module loading order:

modprobe usbcore

modprobe ehci_hcd

modprobe uhci_hcd

Rebooted, and problem solved.

JBoss XA datasources and mbean error

When deploying an application on Jboss that used Oracle and XA datasources, the following error croped up:

org.jboss.deployment.DeploymentException: Trying to install an already registered mbean: jboss.jca:service=OracleXAExceptionFormatter

It took me a while to found out what the issue was, and the lead to the solution was that the error moved from datasource to datasource according to it’s name, namely the xml definition filename.

The issue is quite simple: XA datasources can only have one mbean definition for the OracleXAExceptionFormatter, so choose one, and only one and put the required definition:

<mbean code=”org.jboss.resource.adapter.jdbc.vendor.OracleXAExceptionFormatter”
name=”jboss.jca:service=OracleXAExceptionFormatter”>
<depends optional-attribute-name=”TransactionManagerService”>jboss:service=TransactionManager</depends>
</mbean>

Remove this definition from any other XA datasource and the error should disappear and all datasources can bound now without any errors.