Article

folboteur's picture
article
Reads:

3417

Score:
0
0
 
Comments:

0

Essentials of UserApp JBOSS Clustering -- Quick Details

Author Info

3 November 2010 - 9:57am
Submitted by: folboteur

(View Disclaimer)

Critical JBOSS Clustering Configuration Steps:

Novell's User Application Clustering documentation can be found here.

This note will not duplicate content found there, but highlight relevant details that fit into the procedures described in the Novell documents. These steps were taken in creating both the lab and production cluster environments.

You of course need to choose "All" and establish different engine ID's on each UA install.

Identical Master Keys: Clusters require identical master keys to function, and a key is requested during installation of the UA. During install of the first cluster node, the UA generates a key and places it in the install base folder, in a file called "master-key.txt" .During the second node installation, copy the master-key.txt file to the new server during UA Installation and paste its contents into the UserApp installation..

Add the following to the /opt/novell/idm/start-jboss.sh script on each server, with the Engine ID being different:

This command should be inserted in the "exec" line, replacing the former part that begins with -c, and preserving the part that ends "-b 0.0.0.0"

LAB
 node1
 -c IDM -Djboss.partition.name=Lab_Partition -Djboss.partition.udpGroup=228.3.2.2 -Dcom.novell.afw.wf.engine-id=LabUA3
node2
-c IDM -Djboss.partition.name=Lab_Partition -Djboss.partition.udpGroup=228.3.2.2 -Dcom.novell.afw.wf.engine-id=LabUA1

PROD:  (Note the different group multicast address)
node1   
 -c IDM -Djboss.partition.name=Prod_Partition -Djboss.partition.udpGroup=228.3.2.10 -Dcom.novell.afw.wf.engine-id=IDMUA1
node2
-c IDM -Djboss.partition.name=Prod_Partition -Djboss.partition.udpGroup=228.3.2.10 -Dcom.novell.afw.wf.engine-id=IDMUA2

Modify /opt/novell/idm/jboss/server//deploy/cluster-service.xml file:
" set the mcast_port attribute to 45580 in the lab; 45581 in production.**

**This is only relevant in an environment where you have both a lab and Production (i.e. more than one cluster on the same network)

Of course, all of this must be fronted with a load-balancer of some sort to distribute requests to the different nodes and maintain session/connection information.


Disclaimer: As with everything else at Cool Solutions, this content is definitely not supported by Novell (so don't even think of calling Support if you try something and it blows up).

It was contributed by a community member and is published "as is." It seems to have worked for at least one person, and might work for you. But please be sure to test, test, test before you do anything drastic with it.




User Comments

© 2013 Novell