[mod_cluster-dev] Requests not distributed equally across 2 nodes in a cluster

[Bela Ban]

I have 2 nodes 192.168.1.5:8009 and 192.168.1.9:8009, and I see that the 
requests get distributed unequally: 25% go to node1 and 75% to node2. 
This is shown by lookingh at the elected or read stats below.

I haven't changed anything in the mod-cluster-jboss-beans.xml file. I 
see that we have a DynamicLoadBalanceFactorProvider. Do I need to change 
this to a static one ?


Proxy[0]: [/127.0.0.1:8000]:
Node: [1],Name: node1,Balancer: mycluster,Domain: ,Host: 
192.168.1.5,Port: 8009,Type: ajp,Flushpackets: Off,Flushwait: 
10000,Ping: 10000000,Smax: 1,Ttl: 60000000,Elected: 27747,Read: 
470713,Transfered: 0,Connected: 8,Load: 78
Node: [2],Name: node2,Balancer: mycluster,Domain: ,Host: 
192.168.1.9,Port: 8009,Type: ajp,Flushpackets: Off,Flushwait: 
10000,Ping: 10000000,Smax: 1,Ttl: 60000000,Elected: 68240,Read: 
2131998,Transfered: 0,Connected: 6,Load: 77
Vhost: [1:1:1], Alias: localhost
Vhost: [2:1:2], Alias: localhost
Context: [1:1:1], Context: /, Status: ENABLED
Context: [1:1:2], Context: /invoker, Status: ENABLED
Context: [1:1:3], Context: /web, Status: ENABLED
Context: [1:1:4], Context: /juddi, Status: ENABLED
Context: [1:1:5], Context: /jbossws, Status: ENABLED
Context: [1:1:6], Context: /load-demo, Status: ENABLED
Context: [1:1:7], Context: /jmx-console, Status: ENABLED
Context: [1:1:8], Context: /web-console, Status: ENABLED
Context: [1:1:9], Context: /sun, Status: ENABLED
Context: [2:1:10], Context: /web-console, Status: ENABLED
Context: [2:1:11], Context: /juddi, Status: ENABLED
Context: [2:1:12], Context: /invoker, Status: ENABLED
Context: [2:1:13], Context: /jbossws, Status: ENABLED
Context: [2:1:14], Context: /, Status: ENABLED
Context: [2:1:15], Context: /jmx-console, Status: ENABLED
Context: [2:1:16], Context: /load-demo, Status: ENABLED
Context: [2:1:17], Context: /sun, Status: ENABLED
Context: [2:1:18], Context: /web, Status: ENABLED


-- 
Bela Ban
Lead JGroups / Clustering Team
JBoss - a division of Red Hat