JNDI lookup failed with HANamingService in a cluster

n2 Jun 2, 2005 6:44 AM

Hello,
i hava a cluster with two machines (SUSE Linux Enterprise Server, JBoss 4.0.2). I am testing failover with a Swing client. Everything is working fine when i stop one of the machines. But when i start the JBoss again i get javax.naming.NameNotFoundExceptions while looking up beans which worked before without errors.

jndi.properties:

java.naming.factory.initial=org.jnp.interfaces.NamingContextFactory
java.naming.factory.url.pkgs=org.jboss.naming:org.jnp.interfaces
jnp.discoveryGroup=230.0.0.4
attribute "AutoDiscoveryGroup"
jnp.discoveryPort=1102
jnp.partitionName=JFleet

I run the all configuration. I changed the partition name.
What going wrong?

Daniel.

1. Re: JNDI lookup failed with HANamingService in a cluster

ben.wang Jun 2, 2005 1:41 PM (in response to n2)

Did you see your ejb is deployed correctly on the restart?

-Ben
Actions
2. Re: JNDI lookup failed with HANamingService in a cluster

n2 Jun 3, 2005 3:48 AM (in response to n2)

Yes, all ejbs are deployed on the server after restart.

I deployed the same app on my notebook (SUSE Prof. 9.1, JBoss 4.0.2 cluster on one machine). Everything is fine. So maybe is a problem with two hosts.

I get strange warnings like "Ignoring message from non member cluster1" on both servers from jgroups classes in the log.
Actions
3. Re: JNDI lookup failed with HANamingService in a cluster

n2 Jun 3, 2005 4:23 AM (in response to n2)
The warnings from jgroups are:
WARN [org.jgroups.protocols.pbcast.NAKACK] [JBoss2:32962 (additional data: 17 bytes)] discarded message from non-member jboss1:33049 (additional data: 17 bytes)

Is there a problem with cluster communication which affects the naming service?
Actions

Go to original post