Development Setup


This chapter helps you get an installation of Kitsune up and running.

If you have any problems getting Kitsune running, let us know. See Contact us.

Getting up and running

To get Kitsune running locally all you really need is to have Docker and Docker Compose installed, and follow the following steps.

  1. Fork this repository & clone it to your local machine.

    git clone
  2. Pull base Kitsune Docker images, run collectstatic, create your database, and install node packages.

    make init
    make build

    If you have low bandwidth, you may get a timeout error, see issue#4511 for more information. You can change default pip’s timeout value (which is 60 seconds) by running:

    make build PIP_TIMEOUT=300

    In above command, we are setting default value of PIP_DEFAULT_TIMEOUT to 5 minutes, change it according to your need.

  3. Run Kitsune.

    make run

    This will produce a lot of output (mostly warnings at present). When you see the following the server will be ready:

    web_1              | Starting development server at
    web_1              | Quit the server with CONTROL-C.

    The running instance will be located at http://localhost:8000/ unless you specified otherwise, and the administrative control panel will be at http://localhost:8000/admin/.

Another way you might choose to run the app (step 3 above) is by getting a shell in the container and then manually running the Django dev server from there. This should make frequent restarts of the server a lot faster and easier if you need to do that:

make runshell

The end result of this method should be the same as using make run, but will potentially aid in debugging and act much more like developing without Docker as you may be used to. You should use make runshell here instead of make shell as the latter does not bind port 8000 which you need to be able to load the site.

Run make help to see other helpful commands.

Finally you can run the development server with instance reloading through browser-sync.

npm start

The running instance in this case will be located at http://localhost:3000/.

Further setup

Admin interface

After the above you can do some optional steps if you want to use the admin:

  • Enable the admin control panel

    echo "ENABLE_ADMIN=True" >> .env
  • Create a superuser

    docker-compose exec web ./ createsuperuser
  • Create a profile for this user

    $ ./ shell_plus
    In [1]: u = User.objects.get(username="superuser")
    In [2]: Profile(user=u).save()
  • Log in to the admin panel: http://localhost:8000/admin

Enable development login

If you need to log in as a normal user, add ENABLE_DEV_LOGIN=True to your .env file.

You can create a normal user like so:

docker-compose exec web ./ shell_plus
In [1]: u = User(username="foobar")
In [2]:
In [3]: Profile(user=u).save()

You can then log in as that user by visiting: http://localhost:8000/user/foobar/become

Install Sample Data

We include some sample data to get you started. You can install it by running this command:

docker-compose exec web ./ generatedata

Get AAQ working

  1. Add a product with a slug matching one of the product values in kitsune/questions/ You can do this through the admin interface at /admin/products/product/add/.

    For instance, with a `` value like:
    "name": _lazy("Firefox"),
    "product": "firefox",
    Create a product in the admin interface with its `slug` set to `firefox`.
  2. Add a topic matching a category from that product config, and associate it with the product you just created. You can do this through the admin interface at /admin/products/topic/add/.

    For instance, with a category with a `` value like:
    "topic": "download-and-install",
    Create a topic in the admin interface with its `slug` set to `download-and-install` and its product set to the product you just created.
  3. Finally add an AAQ locale for that product. You can do this through the admin interface at /admin/questions/questionlocale/add/.

Get search working

First, make sure you have run the “Install Sample Data” step above, or have entered data yourself through the admin interface.

  1. Enter into the web container

    docker-compose exec web bash
  2. Build the indicies

    $ ./ es7_init && es7_reindex
  3. Precompile the nunjucks templates

    $ ./ nunjucks_precompile
  4. Now, exit from web’s bash shell

    $ exit


If after running these commands, search doesn’t seem to be working, make sure you’re not running any ad-blocking extensions in your browser. They may be blocking the analytics.js script which search depends on.

Install linting tools

Kitsune uses pre-commit for linting. Install it globally, or in a venv, outside of the docker container with:

$ pip install pre-commit

Then set up its git pre-commit hook:

$ pre-commit install

After this, every time you commit, pre-commit will check your changes for style problems. To run it manually you can use the command:

$ pre-commit run

which will run the checks for only your changes, or if you want to run the lint checks for all files:

$ pre-commit run --all-files

For more details see the pre-commit docs.

Product Details Initialization

One of the packages Kitsune uses, product_details, needs to fetch JSON files containing historical Firefox version data and write them within its package directory. To set this up, run this command to do the initial fetch:

docker-compose exec web ./ update_product_details

Pre-compiling JavaScript Templates

We use nunjucks to render Jinja-style templates for front-end use. These templates get updated from time to time and you will need to pre-compile them to ensure that they render correctly:

docker-compose exec web ./ nunjucks_precompile

Using Django Debug Toolbar

Django Debug Toolbar provides some very useful information when debugging various things in Django, especially slow running SQL queries.

To enable it, ensure DEBUG and USE_DEBUG_TOOLBAR are enabled in .env:


If you are using it to debug slow running SQL queries, you may want to disable the MariaDB query cache. Do that by adding the following lines under the services.mariadb.entrypoint key in docker-compose.yml:

- --query-cache-type=0
- --query-cache-size=0

And restart the database:

docker-compose restart mariadb

Running the tests

Running the test suite is easy:


For more information, see the test documentation.