Not remotely surprised. Any competent engineer knows full well the risk of deploying into us-east-1 (or any “default” region for that matter), as well as the risks of relying on global services whose management or interaction layer only exists in said zone. Unfortunately, us-east-1 is the location most outsourcing firms throw stuff, because they don’t have to support it when it goes pear-shaped (that’s the client’s problem, not theirs).
My refusal to hoard every asset into AWS (let alone put anything of import in us-east-1) has saved me repeatedly in the past. Diversity is the foundation of resiliency, after all.
> as well as the risks of relying on global services whose management or interaction layer only exists in said zone.
Is this well known/documented? I don't have anything on AWS but previously worked for a company that used it fairly heavily. We had everything in EU regions and I never saw any indication/warning that we had a dependency on us-east-1. But I assume we probably did based on the blast radius of today's outage.
My refusal to hoard every asset into AWS (let alone put anything of import in us-east-1) has saved me repeatedly in the past. Diversity is the foundation of resiliency, after all.