This is a big week! Anyone in the world can now go to turboboost.com and create a production Turboboost workspace, completely for free.
Last week this was possible in our dev and beta testing environments, but this week it’s open to the world, as we have finally moved Turboboost to production!
While this isn’t exactly a secret, we haven’t actually told anyone about this because the on-boarding is still a bit rough. That should be ironed out this week, and then we start getting more real-world feedback by asking for feedback from Reddit communities and the larger world.
That said, this hasn’t been without challenges. Our Celery servers are getting swamped running what should be an easy job, the nightly sync of Stripe data of just a few Stripe accounts. So we’re investigating why that is happening, but so far, we are struggling to identify the issue.
For those curious about how we have moved our app to production: we are using multiple AWS accounts through a top-level AWS Management account. So we sign in to the AWS Management org which has no resources itself but allows us to then sign in (via SSO) to any one of our AWS accounts for our dev, beta, or production environment. Each AWS account has it’s own dedicated domain, so the Route53 configs are virtually the same on each.
This approach also easily allows us to limit (for example) junior developers to the ‘dev’ environment or have the testers limited to the ‘beta’ environment. It took a bit of effort and help to set this up properly, but the payoff was almost immediate, and I wouldn’t do it any other way now.
And it makes it easy for us to move our code to increasingly more-robust environments and through multiple stages of testing.
