I cleared the text to save some space but you have given several pieces of information. 1) The first one that needs to be addressed is that you indicate this system was working for ~ 2 years and the problems began about 6 weeks ago and have gotten progressively worse. What changed 6 weeks ago? When you say worse can you qualify what exactly is getting worse? Take an external site that a customer has problems with...make sure you can ping it from one of your internal workstations as you say you have no problems accessing sites with those machines. Then is a customer able to ping that site and not pull up the webpage??? If so...it's MTU. 2) As far as the MTU settings....there is an issue with this if not configured properly. I've seen it posted over and over and over. You should be able to search for it and find some more info using google. The first setting I came across was the following: set network user <name> mtu <number> Don't think that's what you're really looking for. Usually (as do we) you send that information via radius. I'm thinking this may be related to my above statement...if you don't specify it, it defaults to something bad that will cause customer's issues. Typically they won't be able to pull certain sites, the browser will sit there forever trying to pull up certain frames, etc. Have you switched radius servers lately? Is your arc defaulting to a backup server that doesn't have the MTU set correctly? 3) You say the route dies outside your network? How far outside your network? Perhaps something isn't being advertised to your upstream correctly? 4) If you would like me to take a look I don't mind. Create me a temporary account on the arc and I can jump in (without changing anything) and poke around. I would also take a look on the gateway router and see what's going on with it's routes. What changed 6 weeks ago? Todd