JBoss clustering suddenly stopped working.

[mausam-here]

Hello,

We have two jboss clustered (using udp ) on solaris machine and it was working fine for years. Suddenly after recent start the two instances stopped running in cluster. Instead both the instances now run as master with the same PartitionName.

Moreover because of this problem, getHAPartition().callMethodOnPartition .. is taking long time. (Timeout 1 minute.)

I tried the following test

java org.jgroups.tests.McastReceiverTest -mcast_addr 224.10.10.10 [from the earlier master node package]

Then McastSenderTest:

java org.jgroups.tests.McastSenderTest -mcast_addr 224.10.10.10  [from the earlier cluster node package]

and it works fine.

I suspect Unix team might have done to the system which is causing the problem, as I don't see the problem in any of our test environments.

Any suggestions on what are the other tests that I can do (just like above MCastReciever/SenderTest) to prove that there is something wrong with the system configuration?

Best regards,

[galder.zamarreno]

See http://community.jboss.org/message/525159#525159

[bjchip]

Standalone test using multicast ( no JBoss, really simple ) can be built from code here.

http://java.sun.com/docs/books/tutorial/networking/datagrams/broadcasting.html

With this you can verify that you have or haven't got connectivity without involving JBoss at all.   Much much easier to control.

BJ