Trouble is one can't fully escape us-east-1. Many services are centralized there like: S3, Organizations, Route53, Cloudfront, etc. It is THE main region, hence suffering the most outages, and more importantly, the most troubling outages.

This is not true. There are a FEW high-level ones that have their dashboards there, but there's a reason you log into region-specific dashboards

We're mostly deployed on eu-west-1 but still seeing weird STS and IAM failures, likely due to internal AWS dependencies.

Also we use Docker Hub, NPM and a bunch of other services that are hosted by their vendors on us-east-1 so even non AWS customers often can't avoid the blast radius of us-east-1 (though the NPM issue mostly affects devs updating/adding dependencies, our CI builds use our internal mirror)

FYI: 1. AWS IAM mutations all go through us-east-1 before being replicated to other public/commercial regions. Read/List operations should use local regional stacks. I expect you'll see a concept of "home region" give you flexibility on the write path in the future. 2. STS has both global and regional endpoints. Make sure you're setup to use regional endpoints in your clients https://docs.aws.amazon.com/IAM/latest/UserGuide/id_credenti...

us-east-1 was, probably still is, AWS' most massive deployment. Huge percentage of traffic goes through that region. Also, lots of services backhaul to that region, especially S3 and CloudFront. So even if your compute is in a different region (at Tower.dev we use eu-central-1 mostly), outages in us-east-1 can have some halo effect.

This outage seems really to be DynamoDB related, so the blast radius in services affected is going to be big. Seems they're still triaging.

your website loads for a second and then suddenly goes blank. There is one fatal errors from Framer in the console

It is dark. You are likely to be eaten by a grue.

Thanks...probably an issue with Framer during the outage.

Anywhere other than us-east-1 in my experience is rock solid.

Agreed, my company had been entirely on us-east-1 predating my joining ~12 years ago. ~7 years ago, after multiple us-east-1 outages, I moved us to us-west-2 and it has been a lot less bumpy since then.

If you're using AWS then you are most likely using us-east-1 there is no escape. When big problems happen on us-east-1 it affect most of AWS services.

I don't recommend to my clients they use us-east-1. It's the oldest and most prone to outages. I usually always recommend us-east-2 (Ohio) unless they require West Coast.

and if they need West Coast, it's us-west-2. I consider us-west-1 to be a failed region. They don't get some of the new instance types, you can't get three AZs for your VPCs, and they're more expensive than the other US regions.