Trento Telemetry Service

This repo holds the source code of the service Trento uses to collect telemetry data. See the architecture document for additional details.

Deployment

The project offers a terraform deployment option in order to deploy the Trento Telemetry Service. The deployment project deploys the service on AWS ECS (Elastic Containers Service).

The deployment consists of:

• VPC networking layer including all the needed resources
• Amazon load balancer
• Elastic container service based on Fargate (serverless container service)

Requirements

• terraform version >=0.13.x installed
• AWS access and secret keys (with ECS, Load balancer, VPC management, etc IAM rights)
• Already deployed InfluxDB database. The service url, organization id, bucket name and api token with write/right access to this bucket are needed

How to deploy

Follow the next steps to deploy the service:

1. Access the deployment folder
2. Create a new terraform.tfvars file. Use the terraform.tfvars.example as example
3. Authorize the AWS connection setting the access key and secret:

   export AWS_ACCESS_KEY_ID="your-key"
   export AWS_SECRET_ACCESS_KEY="your-secret"
   

4. Create a S3 bucket to store the terraform state files with the name of trento-telemetry-backend
5. In order to use the production system, change to the prod workspace:

   terraform workspace new prod
   # or
   terraform workspace select prod
   

   Use other workspace like test or myname to play with other environments. This creates the S3 object in env/{workspace}/trento-telemetry
6. Run:

   terraform init
   terraform apply
   

7. When the deployment is completed the terraform output shows the url which Trento should send the telemetry data

Grafana

The AWS deployment with terraform/docker-compose deployment offers a Grafana instance to visualize the collected telemetry metrics.

In order to access the dashboards:

• docker-compose: Grafana is available in the port 3000
• AWS: Grafana is available in the port configured by the grafana_port variable (3000 by default) in the same telemetry service URL

Development

Requirements

1. Docker
2. Docker Compose

Environment configuration

Start the platform infrastructure by running make start

It spawns an influx instance and a postgres instance.

Fill .env as needed and Leverage docker compose override features by cp docker-compose.override.yaml.dist docker-compose.override.yaml for customization.