DIDI-SSI-Deploy-issuer

1. Summary

This guide explains how to deploy DIDI Issuer module step by step. To see a more detailed aproach, please refer to OPS Readme.

2. Docker & Node version

Each component is dockerized using Alpine Linux v3.11 with Node v10.x.

3. Prerequisites

3.1. Minimum Requirements

• RAM: 3GB.
• CPU: 1 core.
• HDD: 15 GB.
• OS: Linux (Debian 10 recommended).

IMPORTANT: You also need to have a computer with Linux installed (it could be any distro), so you can install needed dependencies and run deployment script.

3.2. Access Requirements

• SSH access with root user must be enabled.

• Parameter MaxSessions must be disabled (commented) or have a value of 100 (or more).

    ~$ nano /etc/ssh/sshd_config
    
    # Authentication:
  
    #LoginGraceTime 2m
    PermitRootLogin yes
    #StrictModes yes
    #MaxAuthTries 6
    MaxSessions 100
  

• If you update /etc/ssh/sshd_config file, after that, you will need to run service sshd restart to reload the configuration.

• You have to enable passwordless SSH access to server with root using your personal SSH key (you can do it for more than one if you want). For doing that, add your public key to file /root/.ssh/authorized_keys

4. Configuration

4.1. Creating your own environment

1. Clone this repo in your PC (don't do it in the server!).
2. Create a dedicated folder for the environment you want to deploy. To do that make a copy of the folder /ops/example, paste it inside /ops and rename it as you want (in the rest of this document we will call it <environment>).

4.2. Configuring SSH auth for deployment script

1. Copy the same public key/s you used on item 3.2 to folder /ops/<environment>/ansible/ssh-keys and delete the exampleones (before that notice the format your SSH key should have).
2. Edit /ops/<environment>/ansible/custom-vars.jsonto point to the public keys you copied on step 1. Use the example ones to guide yourself, but delete them after you have put the path to your key/s (those are merelly illustrative).
3. Edit /ops/<environment>/ansible/hosts IP with the one from the server you will deploy to.

4.3. Setting up the environment variables

1. Rename the file /ops/<environment>/docker/.env.example to .env
2. Change values of variables inside the file wich are equal to <CHANGE_ME>. Guide yourself using the comments.

IMPORTANT: After you have deployed your environment, you will need to create an Admin user for the Issuer module. To do so (before deploying), you have to set the variable ENABLE_INSECURE_ENDPOINTS = true and follow the procedure explained on item 3 here. If you have deployed a productive environment, after you have created Admin user, for security reasons, you will need to re-deploy Issuer module, but this time with variable ENABLE_INSECURE_ENDPOINTS = false.

To generate some of the variables you will have to run the script /deploy-tools/key-generator.js. For doing that you need to install Node Version Manager. To do so, run the following commands:

~$ curl https://raw.githubusercontent.com/creationix/nvm/master/install.sh | bash
~$ source ~/.profile

To run key-generator.jsrun the following commands inside deploy-tools folder:

~$ nvm install
~$ nvm use
~$ npm i
~$ node key-generator.js

4.4. Setting up MongoDB Initialization script

Change values of variables inside the file /ops/<environment>/configs/mongo-init.js wich are equal to <CHANGE_ME>. Guide yourself using the comments. Use the values of the corresponding variables in .env file.

4.5. Expose MongoDB port

If you are deploying to a testing environment, probably you will want to expose the database port. To do so, edit the ports section of service mongo in file /ops/<environment>/docker/docker-compose-overrides.yml:

• To expose the port: - 27017:${MONGO_PORT}
• To hide the port: - 127.0.0.1:27017:${MONGO_PORT}

5. Deployment

5.1. Configuring Server

If this is the first time you are deploying to the server, you will need to perform tasks like these:

• Install docker.
• Install docker-compose.
• Install other dependencies.
• Enable docker-compose to authenticate using your SSH key, so you can run it over the server but executing it remotely from your computer.

To do this, first of all you need to install Ansible in your computer (see how to do it in your OS here). After that, go inside ops folder and run the following command:

./ops.sh --config-server <environment>

IMPORTANT: You will need to perform this step only if:

• This is the first time you are deploying to your server.
• You need to add an extra authorized SSH Key for ops.sh authentication.

5.2. Deploying solution

To deploy the solution to your server you need to run the following command:

./ops.sh --deploy <environment>

This will pull solution's modules from DIDI's public Docker Repo and will run the corresponding containers in your server.

5.3. Other useful commands

You can deploy only one specific docker-compose module (and its dependent ones) running the following command:

./ops.sh --deploy <environment> <docker_module>

For example:

./ops.sh --deploy testing issuer-backend

Also, you can stop/start/restart one module (and its dependent ones) running the following commands:

./ops.sh --stop <environment> <docker_module>
./ops.sh --start <environment> <docker_module>
./ops.sh --restart <environment> <docker_module>

5.4 Post-Deployment Documentation

Once you deployed everything you need to go to the Post-Deployment Guide where we explain what to do next.