Installation on Heroku

Any questions, problems or suggestions with this guide? Ask a question in our community or contribute the change yourself at https://gitlab.com/baserow/baserow/-/tree/develop/docs .

The Heroku template and one click to deploy button are currently in beta.

Heroku is a platform as a service that enables developer to build, run and operate applications entirely in the cloud. We have created a template that allows you to easily install Baserow on the basic dyno of Heroku. You can also scale up by tweaking some settings.

The template

Deploy to Heroku Beta

The button above can be used to install Baserow on Heroku with one click. Click on it and follow the steps on heroku.com to proceed. After the installation, you can reach Baserow on the URL provided by Heroku. Everything installed via the template runs on the basic dyno of Heroku by default.

Budgeting

On November Nov 28 2022 the hobby-dev (free) Heroku plans were retired.

This means that to deploy Baserow to Heroku, it will require a small cost. We have chosen the cheapest and most secure dynos and addons to deploy with. You can always upgrade your dynos should your installation require more resources (see Scaling below).

  • web ($7/mo): the web formation creates a Basic always-on dyno.
  • heroku-redis:mini ($3/mo): the Redis addon.
  • heroku-postgresql:mini ($5/mo): the Postgresql addon.

Built-in Baserow templates disabled by default

By default, we are using the heroku-postgresql:mini addon because it’s the cheapest ($5/mo, up-to 10k rows) Postgres addon. The Baserow templates require more than 10k rows, so we do not load them by default. If you upgrade your heroku database to support more data and want to load the templates run the following:

  1. Login to Heroku and open your Baserow app
  2. Click “More” in the top right
  3. Click “Run Console”
  4. Enter and run the following command ./baserow.sh backend-cmd manage sync_templates

Every time you upgrade your Heroku Baserow app you will need to repeat the steps above the get the latest Baserow templates.

Could not connect to the API server error

If you are getting a “Could not connect to the API server.” error when logging in or creating an account, then you most likely need to update the BASEROW_PUBLIC_URL config var. You can do so by going to the settings page in your Heroku app dashboard, click on “Reveal Config Vars”, find the BASEROW_PUBLIC_URL and change the value to https://YOUR_APP_NAME.herokuapp.com. Don’t forget to replace YOUR_APP_NAME with the name of your app.

Persistent S3 file storage

By default, the uploaded files are stored inside the dyno for demo purposes. This means that everytime your dyno restarts, you will lose all your uploaded files. Your files can optionally be stored inside an S3 bucket. To do so, you need to add a couple of config vars to the settings. Go to the Settings page and click on “Reveal Config Vars”. Here you need to add the following vars:

  • AWS_ACCESS_KEY_ID: The access key for your AWS account.
  • AWS_SECRET_ACCESS_KEY: The secret key for your AWS account.
  • AWS_STORAGE_BUCKET_NAME: Your Amazon Web Services storage bucket name.
  • AWS_S3_REGION_NAME Optional: Name of the AWS S3 region to use (eg. eu-west-1)
  • AWS_S3_ENDPOINT_URL Optional: Custom S3 URL to use when connecting to S3, including scheme.
  • AWS_S3_CUSTOM_DOMAIN Optional: Your custom domain where the files can be downloaded from.

Non AWS S3 providers

It is also possible to use non AWS, S3 providers like for example Digital Ocean. Below you will find example settings if you want to connect to Digital Ocean Spaces.

Scaling

Even though the template runs on the basic dyno of Heroku by default, it can easily be scaled up to fit your needs. We recommend scaling up if you are going to use Baserow with more than one user simultaneously. You can scale up by changing the dyno type and increase the amount of dynos.

Workers per dyno

To spare resources, every dyno has only one worker by default. If you are upgrading to a standard 2x dyno, you can increase the amount of workers to 2. This can be done via the Config Vars in the Settings. On the settings page, click on “Reveal Config Vars”, find the “BASEROW_AMOUNT_OF_WORKERS” var and set the value to 2.

You can roughly estimate the amount of workers based on the available RAM of your dyno type. Every worker needs around 512MB ram, so a standard x1 dyno should have one worker, a standard x2 can have 2 workers and a performance M can have 4.

Postgres

By default, we are using the heroku-postgresql:mini addon because that supports 10k rows for $5/mo. If you need more rows, you need to upgrade that addon.

Redis

By default, we are using the heroku-redis:mini addon because that addon supports 20 connections for $3/mo. If you are scaling up, you need more connections which means you need to upgrade that addon. In order to roughly estimate how many connections you would need, you can do DYNO COUNT * BASEROW_AMOUNT_OF_WORKERS * 15.

Custom Domain

If you added a custom domain, then you need to change a Config Var on the settings page. Go to the Settings page and click on “Reveal Config Vars”. Here you need to set the BASEROW_PUBLIC_URL value and add your own URL. If your domain is baserow-test.com ,then the value should be https://baserow-test.com. If you don’t have a custom domain then this value can be empty.

Environment variables

  • SECRET_KEY: A unique string that is used to generate secrets.
  • BASEROW_JWT_SIGNING_KEY: A unique string that is used to sign jwt tokens.
  • BASEROW_PUBLIC_URL: The public URL of your Heroku Baserow app. If empty, the default Heroku app URL is used, but if it differs it must be changed. (eg. https://baserow-test.com).
  • BASEROW_AMOUNT_OF_WORKERS: The amount of workers per dyno.
  • AWS_ACCESS_KEY_ID: The spaces API key.
  • AWS_SECRET_ACCESS_KEY: The spaces API secret key.
  • AWS_STORAGE_BUCKET_NAME: The name of your space.
  • AWS_S3_REGION_NAME: Name of the Digital Ocean spaces region (eg. ams3)
  • AWS_S3_ENDPOINT_URL: Custom S3 URL to use when connecting to S3, including scheme.
  • AWS_S3_CUSTOM_DOMAIN: Your custom domain where the files can be downloaded from.

Application builder domains

Baserow has an application builder that allows to deploy an application to a specific domain. Because Heroku has a reverse proxy that routes a domain to the right dyno, the deployed application isn’t automatically available on the chosen domain.

To make this work, you must add a domain alias in the app settings. This can be done by going to the settings of your Baserow app, then scroll to Domains, click on Add domain, and then add the domain you’ve published the application to in Baserow. Make sure that the domain matches the full domain name in Baserow.

Update

When a new version of Baserow has been released, you probably want to update. Unfortunately, it is not possible to update from a public repository via the web interface, so you need to do that via the command line. If don’t have the heroku-cli installed, you can follow the instructions here https://devcenter.heroku.com/articles/heroku-cli. After the cli is installed, you probably need to run heroku login to authenticate with your account.

Make sure that you navigate to an empty directory and then run the commands below. Don’t forgot to replace YOUR_APP_NAME with the name of your Heroku app.

$ git clone --branch master https://gitlab.com/baserow/baserow.git
$ cd baserow
$ git remote add heroku https://git.heroku.com/YOUR_APP_NAME.git
$ git push -f heroku master

The latest version is deployed after the command finishes. You can optionally cleanup the created directory by executing the following commands.

$ cd ../
$ rm -rf baserow