KMS API

Docker Containers

docker build -t kms-api . and docker run --env-file .env -p 8080:8080 kms-api

This is the repository for the Knowledge Management System API. It is currently in the process of being refactored from the old pipeline and contains endpoint for creating, reading, updating, deleting, and querying knowledge objects from the FireStore database.

Setup

API Backend

This codebase requires Python 3.11 or higher and since most default Python installations are 3.9 or lower you may need to install a newer version. When I first built Python 3.11 from source I had some missing dependencies for libraries, so I've included some common libraries below that fixed the issues for me (I don't think all of them are necessary).

sudo apt update
sudo apt upgrade
sudo apt install build-essential python3-dev python-setuptools python3-pip python3-smbus libncursesw5-dev libgdbm-dev libc6-dev zlib1g-dev libsqlite3-dev tk-dev libssl-dev openssl libffi-dev libbz2-dev

Now Python can be compiled from source. You can find all of the releases here. Look for the "Gzipped source tarball" link for the next step. Replace the links and zip/directory name with your version of Python.

wget https://www.python.org/ftp/python/3.11.3/Python-3.11.3.tgz
tar xzvf Python-3.11.3.tgz
cd Python-3.11.3
./configure
make
sudo make install

Clone this repository.

git clone https://github.com/BlockScience/kms-api

Next create a virtual environment and install the required libraries.

cd kms-api
python3.11 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

Run python setup.py to generate your local config file and update settings. Edit config.py to reflect your system configuration, most common change will be PROJECT_PATH which should point to where this repo was clone, most likely your home directory.

For Development

For local testing, you can execute python -m api to run a server on localhost:8000. You can also run python -m generate_key to create an API key for testing.

If you want to run the full stack development server, you will need to set DOMAIN = 'localhost' in config.py, create a self-signed certificate for HTTPS, and set SSL_CERT_PATH and SSL_KEY_PATH in config.py. You can follow the rest of the steps for production, starting at running python setup.py again.

For Production

For production deployment, systemd is configured to run the API as a systemd service via Uvicorn also locally hosted at localhost:8000. Nginx is used as a reverse proxy to forward requests to the Uvicorn server.

First you will need to configure HTTPS using EFF's certbot. Instructions can be found here. Once installed you will want to run the following command:

sudo certbot certonly --nginx

You will receive two paths to a certificate and key which need to be set in config.py as SSL_CERT_PATH and SSL_KEY_PATH respectively. At this point, run python setup.py once again to generate the systemd service file and nginx conf file which will be stored in the conf directory.

Setup systemd service for uvicorn process.

sudo systemctl enable ~/kms-api/conf/kms-api.service
sudo systemctl start kms-api

Install and configure nginx.

sudo apt install nginx
sudo ln -s ~/kms-api/conf/nginx.conf /etc/nginx/sites-enabled/kms-api
sudo nginx -s reload

Making configuration changes

Changes can be made to config.py at anytime. For these changes to affect the production environment, you will need to rerun python setup.py to regenerate the systemd service and nginx conf files. Each service will need to be manually update to reflect these config changes.

To reload the API service:

sudo systemctl daemon-reload
sudo systemctl restart kms-api

To reload the nginx server:

sudo nginx -s reload