Operate enmeshed

Prerequisites

MongoDB

The Connector requires a MongoDB compatible database as its data storage. MongoDB is a document-oriented database. For compatibility and security reasons, the most up-to-date version of MongoDB should be used. For more information, please see https://www.mongodb.com.

If you want to use an open-source database you can use FerretDB instead.

Container Runtime

The Connector requires a Container Runtime like Docker or Kubernetes: Docker is a virtualization technology which introduces highly portable software containers. The Connector is shipped and updated as such a Docker container - the Docker Runtime is the runtime environment which can execute the Docker containers. For compatibility and security reasons, the most up-to-date version of the Docker Runtime should be used. For more information, please see https://www.docker.com.

Visit the official docker docs for installation guides.

Hardware Requirements

No special hardware requirements have been identified so far and as always, hardware requirements strongly correlate with the envisoned usage scenario.

A good starting point for hosting the Docker image of the Connector would be the following:

  • 1 CPU
  • 512MB RAM
  • 1GB HDD

Depending on the usage scenario, higher hardware requirements might be necessary.

Internet Connectivity

A reliable and fast internet connection is mandatory for running the Connector. However, the Connector is only communicating with the Backbone so the corresponding domain (e.g. https://prod.enmeshed.eu) can be whitelisted and the associated certificate can be additionally pinned.

List docker image tags

Read more about listing available docker image tags here.

Familiarize with our policies

Before setting up enmeshed, you should familiarize yourself with our Security Considerations and Privacy Considerations.

Installation with Docker

Make sure that you have installed docker compose. Visit the official installation guide for more information.

ATTENTION: The Docker compose files we provide in this tutorial are not recommended to use in production scenarios. Please read Use Compose in production for more information on how to write a production-grade compose file and our Security Considerations.

Option 1: docker compose including MongoDB

Go through the following steps to start the Connector:

  1. place the file examples/docker-compose-with-mongodb.yml as docker-compose.yml in a folder of your choice
  2. create a config file that can be mounted inside the Connector. Fill the config file using the configuration docs and the example config file. If you used the yml-file from the first step, the connection string looks as follows: 
    mongodb://<db-username>:<db-password>@mongodb:27017
  3. replace all <placeholders> in the compose file with the corresponding values
  4. (optional) follow the steps under log file mounting if you want to persist and access the log files on the host system
  5. execute docker compose up -d in the shell

Option 2: docker compose with existing MongoDB

Visit the official MongoDB website for installation without docker or cloud usage or the docker hub page for information about the installation with docker.

Go through the following steps to start the Connector:

  1. make your existing MongoDB available for the Connector
  2. place the file examples/docker-compose-with-existing-mongodb.yml as docker-compose.yml in a folder of your choice
  3. create a config file that can be mounted inside the Connector. Fill the config file using the configuration docs and the example config file
  4. replace all <placeholders> in the compose file with the corresponding values
  5. (optional) follow the steps under log file mounting if you want to persist and access the log files on the host system
  6. execute docker compose up -d in the shell

Option 3: docker compose with FerretDB

Go through the following steps to start the Connector:

  1. place the file examples/docker-compose-with-ferretdb.yml as docker-compose.yml in a folder of your choice
  2. create a config file that can be mounted inside the Connector. Fill the config file using the configuration docs and the example config file. If you used the yml-file from the first step, the connection string looks as follows: mongodb://ferretdb:27017
  3. replace all <placeholders> in the compose file with the corresponding values
  4. (optional) follow the steps under log file mounting if you want to persist and access the log files on the host system
  5. execute docker compose up -d in the shell

Installation with Kubernetes and Helm

Make sure that you have a running Kubernetes cluster and that you have installed kubectl and Helm.

You have to provide your own MongoDB instance. Visit the MongoDB website for installation without docker or cloud usage or the docker hub page for information about the installation with docker or install it in kubernetes via Helm.

For the installation and configuration head over to the dedicated Connector Helm chart site.

Validate the Connector installation

You can validate the Connector installation by checking its health route. Simply access <connector-baseurl>/health in your browser or using curl.

If the swagger documentation is enabled you can also access it under <connector-baseurl>/docs

Log file mounting

  1. Uncomment the volume mapping in the created docker-compose.yml file
  2. Create a folder where the log files shall be placed. Make sure that the process in the container has write access to the folder e.g. by executing chmod 777 <folder> on your created folder.
  3. replace </folder/of/your/choice> with the path to your created folder

API platform hosted on the (development) Connector

To use the api platform hosted on the Connector you need to make the following config changes:

  1. the http server must be enabled in the configuration of the Connector.

    {
  "infrastructure": {
    "httpServer": {
      "enabled": true,
      "apiKey": "an-api-key"
    }
  }
}

  2. furthermore the API must be activated

    {
  "modules": {
    "coreHttpApi": {
      "docs": {
        "enabled": true
      }
    }
  }
}

  3. the API must not be used in production systems, therefore the tag “debug” must be activated

    {
  "debug": true
}

Troubleshooting

If you encounter any problems while setting up the Connector, head over to the Troubleshooting site.