Hey all - Thanks for taking the time to read this. I just set up my first squid server, acting as a reverse http accel proxy for one of our clusters. The cluster serves up WMV files, at approximately 1-4MB in size. We currently have 2 cacheing servers sitting behind a Foundry ServerIron load balancer, taking a sample of traffic (we're still for now letting some traffic go direct to the webservers as well. The two machines are both dual-processor, one has Athlon MPs at 1800MHz (as reported by /proc/cpuinfo), the other using Opteron 242's at 1600MHz. Both machines have 2GB of ram, and 6 18GB SCSI drives. One of the drives is devoted to the OS, the other 5 are allocated to the squid cache. The problem we're having is that the Squid servers aren't pushing as much traffic as we would expect them to. Each of the machines is only pushing about 100Mb thru 1Gb interfaces, while the apache servers themselves push much more. Looking for any configuration suggestions, tweaking, or anything else that might help. Thanks again, Jeremy
