Load balancing in mod cluster is not working - jboss

Question

I have Jboss server in Linux boxes. And I configured apache server in windows machine. I am able to see all the jboss server nodes in my modcluster manager console.
I have deployed one camel application on all the jboss servers. And I have done the performance test with 2,4,6 nodes. But there is no performance difference.......
Find the jboss configuration
<subsystem xmlns="urn:jboss:domain:modcluster:1.0">
<mod-cluster-config proxy-list="x.x.x.x:6666" advertise="false">
<dynamic-load-provider>
<load-metric type="busyness"/>
</dynamic-load-provider>
</mod-cluster-config>
</subsystem>
For parallel execution of nodes, whether I have to do any other configurations...
Thanks in advance................

Answer 1

1 - download last version of mod_cluster at this link and extract it..
2 - configure your mod_cluster at the httpd.conf file like above..
Listen ##PUT THE BALANCER IP HERE##:80
############### mod_cluster Setting - STARTED ###############
<IfModule ssl_module>
SSLRandomSeed startup builtin
SSLRandomSeed connect builtin
</IfModule>
# MOD_CLUSTER_ADDS
# Adjust to you hostname and subnet.
<IfModule manager_module>
Listen ##PUT THE BALANCER IP HERE##:6666
ManagerBalancerName mycluster
<VirtualHost ##PUT THE MACHINE IP HERE##:6666>
<Location />
Order deny,allow
Deny from all
Allow from 192.168.0
</Location>
KeepAliveTimeout 300
MaxKeepAliveRequests 0
AdvertiseFrequency 5
EnableMCPMReceive
<Location /mod_cluster_manager>
SetHandler mod_cluster-manager
Order deny,allow
Deny from all
Allow from 192.168.0
</Location>
</VirtualHost>
</IfModule>
############### mod_cluster Setting - ENDED ###############
3 - Set each of your jboss node's name
<server name="node1" xmlns="urn:jboss:domain:1.2">
4 - Add the instance-id attribute in web subsystem as shown below in both the standalone nodes
<subsystem xmlns="urn:jboss:domain:web:1.1" instance-id="${jboss.node.name}" default-virtual-server="default-host" native="false">
<connector name="http" protocol="HTTP/1.1" scheme="http" socket-binding="http"/>
<connector name="ajp" protocol="AJP/1.3" scheme="http" socket-binding="ajp"/>
.
.
.
</subsystem>
5 - Add the proxy-list in the attribute in mod-cluster-config of modcluster subsystem, which would be having IP Address and Port on which your Apache server (the balancer) is running so that JBoss server can communicate with it, as shown below in both the standalone nodes
<subsystem xmlns="urn:jboss:domain:modcluster:1.0">
<mod-cluster-config advertise-socket="modcluster" proxy-list="##PUT THE BALANCER IP HERE##:80">
.
.
.
</mod-cluster-config>
</subsystem>
6 - Now you can go to http://BALANCER_IP:80 and test it and to manage the jboss instances with mod_cluster go to http://BALANCER_IP:6666/mod_cluster_manager
**Obs: if you want to run jboss in standalone mode you CANNOT use the "-b" flag with the ip 0.0.0.0 that listens requests from all IPs.. I recommend you use the IP of the machine that's running the jboss itself

Answer 2

with sticky-session="true" (default), balancer keeps sending requests to the particular node to whom the session belongs as long as it is healthy.
If you tell me how did you test, especially: how many clients vs. how many requests, or not etc., I will be able to help you.
Furthermore, consider editing capacity attribute of load-metric element.
BTW: "busyness" considers threads in thread pool being occupied with serving requests. You might find that this is not the bottleneck of your system. You might want to add heap, requests or other metrics. See http://docs.jboss.org/mod_cluster/1.2.0/html_single/