Wired mag article on spammers playing traceroute games with trojaned boxes

Joe Abley jabley at isc.org
Thu Oct 9 16:53:30 UTC 2003



On 9 Oct 2003, at 12:19, Vinny Abello wrote:

> Personally, I think preventing residential broadband customers from 
> hosting servers would limit a lot of that. I'm not saying that IS the 
> solution. Whether or not that's the right thing to do in all 
> circumstances for each ISP is a long standing debate that surfaces 
> here from time to time. Same as allowing people to host mail servers 
> on cable modems or even allowing them to access mail servers other 
> than the ISP's.

"Hosting a server" looks very similar to "using an ftp client in active 
mode", "playing games over the network" or "using a SIP phone" to the 
network. Enumerating all permissible "servers" and denying all 
prohibited ones arguably requires an unreasonable shift of intelligence 
into the network. Allowing inbound connections by default and blocking 
specific types of traffic reactively has been demonstrated not to be an 
adequate solution, I think.

A more aggressive policy of blocking all inbound connections (and 
analogues using connectionless protocols) essentially denies direct 
access between edge devices, which implies quite an architectural 
shift.

I think it's more complicated than "prevent residential users from 
hosting servers".


Joe




More information about the NANOG mailing list