I have a distributed worker.
2 instances in each region: iad, lhr, hkg
About half an hour ago the workers from iad & lhr can no longer connect to the DB.
The ones from hkg can.
Wondering if anyone ever seen anything like that before.
Hello!
Last night, cloudflare had broken routing to our north american IPs. By any chance, your DB’s IP was a cloudflare owned IP?
Also, are you still facing this?
Akshit
Fly.io Infrastructure Operations
Hi Akshit,
Nope. It behind Amazon Route 53.
Both N.America and Europe were affected (I checked at least iad, lhr & ams) - meaning machines on fly.io with static egress-ip-s from these regions couldn’t connect to my lhr DB.
Resolved within 30-45 minutes.
Cloudflare has no incidents at that time as well.
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.