...if you fuck up with the routing protocol BGP it can cause massive problems or it is always DNS.
I saw where someone shared Cloud Fare's summation of the incident here on Poal last night . Definitely BGP related.
Convenient timing. I used to work for a webhosting company, major changes - ESPECIALLY shit on the routers - always were scheduled for 2-3am. Backups of data were scheduled for that timeframe too as you want a time of low activity so your file traffic to the NAS (in our case the tape robot) doesn't stomp your internal network between the servers to death.
I think they used the downtime to delete shit that was talked about by the whistleblower probably. What was talked about?
(post is archived)