Often my first visit of the day I often have to relog back in because the cookie has gone away, but apart from the sql server outage issues you mentioned, its all fine to here. I guessed the cookie going away was down to the hosting provider doing some kind of reboot out of hours or something.
My packets take the atlas.cogentco.com transit except for the last few hops which is in the oneandone host network so a different route completely than the uk people. That **** item is probably the hosting companies load balancer, and configured to ignore icmp requests for security reasons...
If it is a broken routing table between the uk and the us, theres not a lot brass machine can do about that though. And mostly regardless of isp in the uk, you take the same routes because theyre all either reselling bt broadband kit, or transiting on their backbone connections because of peering arrangements. But, theres I'd imagine, already a load of angry emails flying round between companies technicians and posts to the isp mailing lists about the broken peering if so as it will have been picked up by monitoring tools, and it will get fixed because everyone has service level agreements to meet or suffer a contractual financial penalty, and lost money talks really loudly
