I am attempting to run a Django application using the `runserver` management command in a pipelines build machine. I am doing this so that I can perform some testing that requires a running django instance.
When I run the runserver command and tell it to run on `0.0.0.0:8000` it does not show the normal startup information, and if I run it with nohup and then try to curl `0.0.0.0:8000` on the build machine I get the error `0curl: (7) Failed to connect to 0.0.0.0 port 8000: Connection refused`.
I can successfully get the django app running with the same commands in a local version of the docker image the build machine uses, so it seems to be something specific to running on the build machine. I have also tried switching up the port it is running on to see if that resolves it but to no avail.
My pipeline looks like:
feature/automate-ui-tests:
- step:
name: Run UI tests
caches:
- node
- pip
script:
- apt-get update && apt-get install -y firefox chromium-browser libcurl4-openssl-dev libssl-dev
- pip3 install virtualenv
- virtualenv -p python3 venv
- . venv/bin/activate
- pip install -r app/test-requirements.txt
- export TZ='Pacific/Auckland'
- nvm install
- npm install
- nohup npm run start &
- python app/manage.py migrate --verbosity=3 --settings=app.settings.test-ui
- python app/manage.py createcachetable --verbosity=3 --settings=app.settings.test-ui
- nohup python app/manage.py runserver 0.0.0.0:8000 --verbosity=3 --settings=app.settings.test-ui & # this step is where I have an issue.
- npm run test-ui -- all -r spec,xunit:./app/test-reports/testcafe.xml
services:
- postgres
- redis
Any thoughts? The other management commands like `migrate` and `createcachetable` work fine so it seems to be specific to `runserver`.
Cheers,
