Operations

Overview

The application is setup with the following tools:

• Terraform - for creating infrastructure in AWS for each environment
• Docker - to build and run application components
• Docker Compose - to connect the application components together and deploy them as a stack
• make - to simplify operations tasks
• Prometheus - to gather metrics from application host and Docker containers
• Grafana - to visualize metrics gathered from Prometheus and handle alerting

Environments

The application is hosted on AWS, there are several application environments, each of them is described with Terraform in src/terraform/main.tf file. Terraform is executed manually, in order to add/modify/delete an environment, modify the code and run terraform plan to see the changes and terraform apply to execute them.

Each environment consists of:

• VPC network (with subnets, route tables, IGW)
• Security Groups
• EC2 instance
• Elastic IPs associated with EC2 instance
• Route 53 record (only for environments using govtool.byron.network domain)

For each environment, the frontend is hosted at root and the backend is at /api.

List of public environments

beta

A beta environment connected to sanchonet Cardano network.

Available at https://sanchogov.tools/. The DNS record for this domain is created manually.

Deployment

Deployment is performed via GitHub Actions workflow (.github/workflows/build-and-deploy.yml).

The workflow performs the following steps:

• check if the environment is defined in Terraform (to avoid deployment attempt to inexistant environment)
• build of frontend app
• build of backend app
• generate configuration files and upload them (over SSH) to the target environment
• setup the application compoments with Docker Compose on the target environment

The workflow can be triggered directly from GitHub Actions panel. When ruuning the workflow, you need to specify:

• Cardano network to be used
• environment name
• optionally skip the build process (frontend and backend) - useful when there are plain configuration changes that do not require the application to be rebuild

Monitoring

Monitoring is achieved with Prometheus and Grafana, which are deployed together with each environment. Grafana is available at /grafana, the admin password is defined in a GitHub Actions secret GRAFANA_ADMIN_PASSWORD.

Each Grafana instance is managed as code and provisioned with YAML files located at src/config/grafana-provisioning. This includes a default datasource, dashboard and alerting. The alerts are configured to be sent to Slack via Slack bot.

The Slack bot OAuth token and recipient are defined with GitHub Actions secrets GRAFANA_SLACK_OAUTH_TOKEN and GRAFANA_SLACK_RECIPIENT, respectively.

Frontend

Deploying new versions of the application is done using Github actions

1. Open repository
2. Select "Actions" tab
3. From left menu choose "Build and deploy app"
4. From the droping options - "Run workflow", select the branch, Cardano network and type of environment for your deployment
5. Press "Run workflow"
6. Wait for the final effect. It's done.