Might it be an idea to maintain a map of existing connections to avoid this eventuality? Yes, I'm volunteering :) Feel free to send me your setup and I'll try to hack out a graphical map. Unless it's already been done, of course. Sampsa On 11 Nov 2009, at 16:21, Johnny Billquist wrote: And to be clear. When you talked LAT with London, the traffic actually passed through Uppsala, Sweden, on the way, since just about every bridge program that I know is using Uppsala as the other endpoint. This is not really neccesary, although having just one hop might be a good point in itself. But the bridge program can act as a chain of hops as well. The only restriction is that it don't have anything like STP, so if you create a loop, you are in big trouble. Johnny Steve Davidson wrote: In past I have used LAT (from NH, USA) to connect to CHIMPY (London, UK) and MIM. Johnny's bridge program is running on a NetBSD system between my DEClab LAN and MIM on the other end. It works just fine! -Steve -----Original Message----- From: owner-hecnet at Update.UU.SE on behalf of Sampsa Laine Sent: Tue 11/10/2009 10:23 To: hecnet at Update.UU.SE Subject: Re: [HECnet] Talking of other possible protocols But Johnny's bridge does however support bridging it so in practice it can be used for long distance connections on HECnet... Sampsa On 10 Nov 2009, at 15:21, Paul Koning wrote: Yes, LAT is a layer 2 protocol so it's not routable, and its timers and delay assumptions are for LANs, not for long distance networks. paul -----Original Message----- From: owner-hecnet at Update.UU.SE [mailto:owner-hecnet at Update.UU.SE] On Behalf Of Mark Wickens Sent: Tuesday, November 10, 2009 10:19 AM To: hecnet at Update.UU.SE Subject: [HECnet] Talking of other possible protocols Am I right in thinking that LAT is only suitable for the local network? I just tried Local> connect pdxvax from the DECserver 90M just for a laugh. Regards, Mark.