Open source implementation of ETSI ISG CIM standard known as NGSI-LD. An information model, representation format and open API intended to make it easier for end-users, IoT devices, open data sources and 3rd-party applications to exchange information.
- Node.js (https://nodejs.org)
- MongoDB (https://www.mongodb.com)
Before starting the project, you have to install the node dependencies. To do so, run:
npm install At this step, you will create indexes on collections and default user before starting the server. Before you run the following command, you have to start mongo service and create a database. As described in config file (/config/config.js), we start locally a mongo service and we create 'ngsi_ld_bd' database.
npm run migrateTo start server, run:
npm startA server will start locally listenning on port 3000. The port number is configured in config file (/config/config.js).
Authentication mechanism based on jwt. The authentication endpoint is localhost:3000/login. To retrieve a JSON Web Token from a default user (admin/admin4djane), valid for an hour by default, run the following query:
curl -d '{"username":"admin", "password":"admin4djane"}' -H 'Content-Type: application/json' -X POST "http://localhost:3000/login"The user credentials are configured in config file (/config/config.js).
To test that you have correctly installed the server, you can run your first query
curl -H 'Accept: application/ld+json' -H 'X-AUTH-TOKEN: insert_genereted_jwt' -X GET "http://localhost:3000/ngsi-ld/v1/entities"This will return an empty set, since nothing has been created yet. Note that you must specify the accept header, otherwise you get an error. Look here for more information about the API.
Here is a simple example that will get you up and running with docker. It is easy to build, run and you can even deploy the images to your own repository if you like. This is not meant to be a docker tutorial, but just a basic introduction to get djane runnnig.
Building the docker file follows the usual docker build command
docker build . -t insert_name_here/djane:1.0.0Of course you should replace the tag with whatever you prefer.
If you already have a running mongo database on your local machine, it is easiest to make djane run on your hosts network. If not just follow these steps to get a nice and isolated instance up and running.
docker network create djanenetThat will create the docker network which we will run our mongodb on and our djane service on.
docker run -d -v $PWD/data:/data/db --name mongodb --network djanenet mongoNotice the port is not mounted and that data will be stored on your local machine in a data folder. According to the official documentation here running host mounted volumes on Windows and OSX is not officially supported, so if you run into problems follow the documentation.
docker run --rm -it --network djanenet -e DB_HOST=mongodb insert_name_here/djane:1.0.0 node migrateThis step should run the migrate command against your mongodb.
docker run -d --network djanenet -p 3000:3000 -e DB_HOST=mongodb insert_name_here/djane:1.0.0Notice here that we are setting the environment variable DB_HOST to mongodb. This is the hostname that the mongodb will have in our example. If you are running against a different mongodb, then use the IP or hostname that makes sense for you. Check the Dockerfile for other configurable values.
docker run -d --network host -p 3000:3000 insert_name_here/djane:1.0.0The final example above is in the case that you are running your mongodb natively on your localhost machine.
Using docker-compose is not recommended for a production environment, and the services launched are only for testing and evaluation purposes.
- Docker Compose (https://docs.docker.com/compose/install/)
A YAML file, namely 'docker-compose.yml', is defined to configure the application’s services. This defines two dockers containers:
- djaneio which represents the NGSI-LD service.
- mongo which represents the persistence layer.
To launch the services execute the command below:
docker-compose up