I've been testing a new Apache and Tomcat configuration and have run into a situation where Apache will enter a tight loop. The situation does not correct itself even when no traffic is being sent to the Apache web server. When I truss the process, I see the following: /17: connect(44, 0x01C68EE8, 16, SOV_DEFAULT) Err#133 EISCONN I'm trying to test for a high load web application server, so my setup is as follows: OS - Solaris 10 3 Apache 2.2.15 instances using worker MPM model running as Solaris services on ports 80, 81, and 82, each configured for 1024 MaxClients. 6 Tomcat 6.0.28 instances listening only for AJP/1.3 connections, 512 maxThreads The Apache web servers connect to 6 tomcat instances using a JK connector set up to load balance across the 6 tomcats Java 1.6 Basically we see this condition get hit after sending some high load to the web server. Once the load stops, several of the httpd processes are maxing out several CPU cores yielding the truss output shown above. Any help would be much appreciated. Thanks in advance. -George