AWS management APIs in the us-east-1 region is what is affected. I'm guessing Heroku uses at least the S3 APIs when deploying new versions, and those are faili

AWS management APIs in the us-east-1 region is what is affected. I'm guessing Heroku uses at least the S3 APIs when deploying new versions, and those are failing (intermittently?). I advise not touching your Heroku setup right now. Even something like trying to restart a dyno might mean it doesn't come back since the slug is probably stored on S3 and that will fail.