Notes for Server Contributors

It is recommended to run the project with python 3.11.

Starting the Service

The different notebooks in ./notebooks provide more complete documentation regarding the set up and administration of the server.

On local machine

To start the container on a local machine, first create a mondodb volume with docker volume create mongodata, then go to lomas/server/ and run docker compose up. If you encounter any issue, you might want to run docker compose down first.

Additional services

Running docker compose up will also start two additional services automatically:

a jupyter notebook environment that will be available at the address http://127.0.0.1:8888/ to interact as a user with the server
a streamlit application that will be available at the address http://localhost:8501/ to interact with the server and the administration database as an administrator.

On a kubernetes cluster

To start the server on a kubernetes cluster, first add the repo with: helm repo add lomas https://dscc-admin-ch.github.io/helm-charts and then install the chart with: helm install lomas-sever lomas/lomas-server

On Onyxia

To start the server on Onyxia, select the lomas service, (optionnally adapt the administration and runtime parameters) and click on ‘Lancer’.

Tests

It is possible to test the server without an mongodb administration database. One can do so by executing the run_basic_tests.sh script. Make sure to have an activated python venv in a linux environment with the server requirements installed for it to work.

It is also possible to test the server with standard tests or integration tests (for the mongodb). The run_tests_and_converage.sh script runs the integration tests. Again, make sure to have an activated python venv in a linux environment with the server requirements installed for it to work.

Local tests can also (except for those using the mongodb_admin) can be run with a simple python -m unittest discover -s . from the lomas_server directory. The tests will be based on the config in lomas/server/lomas_server/tests/test_configs/test_config.yaml and be executed with the AdminYamlDatabase.

Tests are also automatically run in GitHub workflows during different events (pull requests, pushes, etc.). Please refer to the general notes for contributors for more informations.