Thank you! CloudFlare's response is appropriate. The incident itself and lack of response (for HOURS) from Verizon's side is absolutely unacceptable. It's 2019

Thank you! CloudFlare's response is appropriate. The incident itself and lack of response (for HOURS) from Verizon's side is absolutely unacceptable. It's 2019, filtering ALL of your customer's routes according to - at least - the IRR (including the legacy ones connected to the old router in the closet) and having a responsive 24/7 NOC contact in PeeringDB are a matter of course. Proper carriers like NTT go above and beyond simple IRR filtering nowadays with things like peerlock (http://instituut.net/~job/peerlock_manual.pdf http://instituut.net/~job/peerlock_manual.pdf). AT&T uses RPKI and was completely unaffected: https://twitter.com/Jerome_UZ/status/1143276134907305984 https://twitter.com/Jerome_UZ/status/1143276134907305984