Getting Started

• Create a new DO instance (whatever size, need to load test this)
• Setup ssh keys for the instance
• Clone https://github.com/nushackers/hnr-infra
• In a separate directory, clone https://github.com/nushackers/hnr-api
• Change the domain names and relative directory of api if needed.
• Run the ./deploy-up.sh to start all the services, and ./deploy-down.sh to stop everything

Droplet Components

• Traefik + Let’s Encrypt
  • This is the main proxy where all the traffic comes through and gets directed to different containers.
  • See config for additional configurations on this
• Prometheus
  • Where all the logs are aggregated. This is connected to an external Grafana instance for monitoring
  • See prometheus.yml for the different services we’re monitoring
• hnr-api
  • This is where out API is deployed

Observability Tools

• Traefik Metrics
• Node Exporter
• prometheus-net

Grafana Dashboards

• We have a couple of Grafana dashboards to visualize the time series data we can get from Prometheus. Tentatively, we have three main dashboards
• Node Exporter - This gives us info on system details: RAM, CPU Usage, Networking
• Traefik - This gives us overall incoming and outgoing traffic information
• ASP.NET Core Dashboard - This gives us detailed per controller information on the API calls - This isn’t very detailed, but is the best template I can find. Probably need to construct our own dashboard for this