We had only set up squid on one server. And that server currently is serving the same images using apache. We're still playing with it and see if we can migrate the majority of our images to use squid. I'll upload the config settings asap. The number of network connections tell me 1 of 3 scenarios #1) Abuse/denial of service (like you suggested) - but since the web daemon model works fine, i have a tough time believing it. Remember one person's Denial of Service attack is nothing more than a normal day for zorpia's load :) #2) misconfigured squid that somehow is not closing sockets as soon as it uses them , so it spirals out of control #3) misconfigured squid not efficient and causing trouble under hte high load I think its most likley #2 or #3, which again brings us back to the 1800 config options to learn about on squid. The config i had works great for about 1/2 your normal load but exploads once full load hits it. I do believe configs in the squid.conf is what is the issue, since cahce log doesn't say there is trouble. I dont see why that box would have trouble keeping up since it works fine in apache only mode... On 7/12/05, Robert Borkowski <rborkows@xxxxxxxxxxx> wrote: > Jeffrey Ng wrote: > > hello, my site, zorpia.com, is a pretty popular site. we get more than > > a million pageviews per day. and we are trying to use squid server to > > serve only navigation images like the tab images or buttons. i don't > > think anybody will abuse our navigation images. i think the hits are > > caused by the huge amount of users we have. is our situation normal to > > you considering there are no abuse? > > How many squid servers are there, and how are they loadbalanced? > Can you post your squid config and output of squidclient mgr:info ? > > -- > Robert Borkowski >
