I am using Apache with mod_jk 1.2.10(prefork MPM) running against Tomcat 5 within JBoss on Linux using JDK 1.4.2.07 Reading thru some of the postings within this Forum, it looks like we would expect to see as many threads in tomcat as the number of MaxClients defined in Apache. In our setup, we have a setup of 8 webservers with each configured to handle MaxClients=625. On the Tomcat side we have 8 servers with each setup to handle 400 MaxThreads. We allow in < 600 simultaneous requests into the webserver farm using a hardware loadbalancer. But when i check via netstat on the servers hosting the Apache ws, i see in addition to established connections on port 8009 to Tomcat, there are a large number of connections in CLOSE_WAIT and TIME_WAIT state. Does this mean that to take care of N simultaneous requests, I would need to configure MaxClients > N to accomodate connections which could be busy due to CLOSE_WAIT and TIME_WAIT? If so, by what factor? I am not sure how long these sockets take to get release to the next request coming in from Apache. My goal is to reduce the number of Tomcat threads running within JBoss and rather than timing out connections within Tomcat, i would prefer to have MaxClients = MaxThreads.
Thanks --------------------------------------------------------------------- To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
