This week's book giveaway is in the OO, Patterns, UML and Refactoring forum. We're giving away four copies of Refactoring for Software Design Smells: Managing Technical Debt and have Girish Suryanarayana, Ganesh Samarthyam & Tushar Sharma on-line! See this thread for details.
I have two JBoss 5.1.1 servers on separate machines that I have configured as one cluster.
I've deployed a simple counting bean to both servers and the clustering seems to work fine.
When I shutdown one of the servers I see all the requests routing to the remaining server.
However, when I restart the server the cluster doesn't reform, and subsequent requests
still goes to just the one server.
Annoyingly, this worked fine two days ago and I successfully shut down server one then server
two with both rejoining the cluster and messages routing as they should, either to one server
when one was down or to both when both were up. I went home happy. But next day, the
restarted server was not rejoining the cluster.
I've followed all the advice I can find, setting the shun parameters etc.
Given that it worked fine a copuple of days ago, I suspect that it may be a timing issue (e.g.
the network may be just a tad slower during the day than in the early evening).
Does anyone have any suggestions?
for TCP Clustering did you set TCPPING attributes correctly?
Which JBoss Version?
Joined: May 10, 2012
Hi Saahil, thanks for responding.
I'm using JBoss 5.1.1. I don't know if it's using TCP or UDP Clustering. How do you tell?
I'm using the 'all' configuration and followed out-of-the-box instructions, so whatever that is using is what I'm using.
Excuse my ignorance, I'm still new to JBoss.
Joined: Dec 26, 2011
JBoss uses UDP clustering as the default cluster configuration. Communication among jboss nodes is done using Jgroups protocol. For setting up the cluster you should be knowing the configuration parameters needed to setup.
Thanks for your reply. We actually ended up pseudo-clustering using load balancers and now we are redeploying to others server types so the issue kind of resolved itself over time.
But this info may well be useful for other applications.
I’ve looked at a lot of different solutions, and in my humble opinion Aspose is the way to go. Here’s the link: http://aspose.com