"There is a problem with one of our routers downstairs, that Anthony is already working on, and should be back up in a few minutes. ETA is under 30 for the entire problem, but most likely will be resolved much quicker than that."
Then from Anthony:
"While attempting to modify an access-list to block malicious traffic destined towards vulnerable cPanel servers a CPU spike occured on our primary MSFC2; this caused a BGP Hold Time Expiration which in turned resulted in the reconvergence of our BGP announcements and routes.
We apologize for any inconvenience this may have caused and all connectivity should be fully restored at this time."
I agree totally with you. Seems there was a mad rush to get things updated. I had problems earlier with one of these servers trying to make that update. It is mentioned that there are a flood of requests attempting to find the exploit.
Due to the current mass-migration of cPanel, DedicatedNOW was forced to physically reroute portions of the network to accomodate the extra traffic. During this process we experienced several issues with a spanning-tree loop and CPU spike following an access-list modification. These issues caused our BGP Hold Timer to expire with NAC, resulting in a reconvergance of our routes and announcements. We are currently working with cPanel to access their network equipment and ensure that there will be no additional network issues.
Network connectivity is currently restored, but obviously, due to the continued updates at numerous datacenters their actual servers may become saturated, resulting in slower than usual performance and updates.
If DedicatedNOW/Pegasus customers experience any further issues we ask that you submit a ticket with [email protected] so we may promptly investigate.