What's New

You can now upload custom SSL certificates for your applications. We'll upload these to your web servers & load balancers and configure Caddy to use them for serving requests to your app. If you don't want to use LetsEncrypt for some reason, this will make it easy for you to still use SSL. 

We now use the AWS API to look up Regions and Instance Types available to your account. The regions and instance types will always be accurate and list what's available for you to use.

We've just updated deployments to set a few environment variables on each deploy.

We also now write a REVISION file with the git commit sha as well for easy access.

You can now backup your unmanaged (local) Redis database with Hatchbox. Just like PostgreSQL and MySQL backups, you can configure how often they run and where they get uploaded by visiting the database backup settings.

This will backup the entire Redis instance, so you will only need to enable backups on one Redis database in Hatchbox. Redis doesn't treat database separately like other databases.

You can now enable socket activation for processes like Puma.

Why is this important? When HTTP requests hit a server, Caddy will forward the request along to your web server like Puma. Puma is listening on socket and that's how Caddy talks to Puma. During a restart, Puma will stop listening to the socket for a short period of time. Caddy will not receive a response from Puma and will render a 500 error.

With Socket Activation, Systemd will keep the socket open and buffer requests while Puma restarts. During a restart, no requests will be dropped and will simply pause until Puma is ready to start handling requests again. 

AWS changes EC2 instance IP addresses anytime the server is stopped. Originally, we solved this by adding Elastic IPs to the servers, but AWS limits the amount of Elastic IPs you can have.

We've replaced this with a script that hits Hatchbox's API with the current IP address each time your EC2 instance boots. This solves the problem elegantly and allows us to remove the need for Elastic IPs. 

We've rewritten Hatchbox from scratch to bring you a huge amount of improvements. 

Here's a quick list of some major improvements:

Plus, there's more improvements than we can even remember. It's been a long time coming.

What about Hatchbox Classic?
Don't worry! Hatchbox Classic isn't going anywhere. We'll keep it running and do routine maintenance so your applications are safe and sound. However, we won't be adding features to it anymore. For example, we won't be adding support for Ubuntu 22.04 Jammy on Classic.

You can login to Hatchbox Classic at https://classic.hatchbox.io

If you'd like to upgrade to the new Hatchbox, we recommend the following steps:

This will provide you a smooth upgrade process with minimal downtime. Since we've upgraded all the underlying tools in Hatchbox, it's not possible to upgrade Classic servers to the new configuration easily. Plus, this way you'll be able to keep the old version running while you test out the new version.

You can now start a backup manually for an unmanaged database on the Web. Just click "Start Backup" on the database page and it will kick off a job to start a manual database backup whenever you want.

We now support deploying App from a subfolder. This is useful if your app codebase resides in your repository's subfolder.

You can find the setting under App > Repository. 

You can now rename* Apps from the settings page to better suit your needs. 

renaming an app may result in temporary downtime

We've upgraded Hatchbox to use ed25519 SSH keys by default. SSH RSA is no longer accepted by default in Ubuntu 22.04 and we've adjusted accordingly to support them.

For clusters created before April 2022, we've generated new SSH keys and have added them to your servers. There's nothing you need to do on your end. 🎉

To connect to your servers over SSH, make sure you're using a compatible SSH key for Ubuntu 22.04. If you don't, SSH will refuse the connection.

You can now deploy to Ubuntu 22.04 LTS Jammy Jellyfish servers. 22.04 is also the new default version used for any new servers that are created. 

Long term support means this server version is supported for 5 years and is the best version to use for production servers.

Hatchbox will now detect the Bundler version from your application's Gemfile.lock and install that during deploy. This will make it easier to deploy older Ruby applications that use Bundler 1.x.

New versions of Ruby have just dropped with important security updates. Check out the release for more details: https://www.ruby-lang.org/en/news/2022/04/12/ruby-3-1-2-released/

We've updated our AWS VPC creation to only allow ingress on ports 22, 80, and 443 by default. 

Previously, we allowed all ports which is a less secure default. If you created a VPC with Hatchbox v2 before April 10th, we encourage you to update your VPC's ingress ports and restrict them.

You can now add domains to an App through the API. This will add the domain and update Caddy to route the domain to your app. It will also automatically provision SSL certificates with LetsEncrypt when the first request comes in using the domain.

Check out the changes here: https://www.ruby-lang.org/en/news/2022/02/18/ruby-3-1-1-released/

You can now transfer your cluster to another user. 

Go to the settings page of your cluster. 

You can now provision and deploy to Amazon EC2 instances! 

We're working on Amazon RDS and ElastiCache integrations for provisioning AWS managed databases next.

You can now deploy apps from GitLab or BitBucket by connecting your account and selecting it as the git host for your application.