On Fri, Apr 09, 1999 at 09:39:18AM -0700, Scott Robertson wrote:
Exodus claims that they are aware of the problem and our in the process of upgrading the bandwidth to their various peering points. Same old story. They've requested that I send them traceroutes from the troubled connections. I figured we could have some fun with this, if the zope list is willing to help. If anybody thinks that the zope.org is running slow could you please e-mail a traceroute to traceroute@codeit.com.
I don't usually admit to having worked for Sprintlink during the early days, but ... oh well, there you have it :-) Anyway, the point Michel is making is completely accurate, what a hosting company needs is private peering arrangements with SLAs that guarantee some accountability. When I was at Sprint we were VERY agressive in pursuing private peerings (ANS/UUNet/MCI at the time, everyone else was <1%) and making sure that they were at least sized reasonably. You can control these. MAE* is a joke, nothing more, it'd be like claiming you had a CIX connection in 1995, interesting perhaps, but not terribly useful for getting traffic somewhere. Chris -- | Christopher Petrilli ``Television is bubble-gum for | petrilli@amber.org the mind.''-Frank Lloyd Wright