Is the only available JK2 load balancing algorithm a weighted round robin? I would like to switch to one that has a primary server and then switches to round robin to reach the secondary server when an error is thrown. The reason I would like to do this is because we have 5 web servers each with a apache/tomcat pair fronted by a Cisco Local Director. I would like the request to stay on the machine that cisco sent the request to as long as the local tomcat is accepting requests. I currely have annecdotal evidence that the current weight round robin approach is a little flakey and is sending WebServer01 more requests then are the server's fair share. I wrote a logging filter and some RequestHeader statements in the http.conf to try to get hard evidence that this is happening.
Thanks, --Angus --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
