XMRig Mining Monitoring with Grafana and Prometheus

Credits to Dockprom https://github.com/stefanprodan/dockprom

How this works

The basic idea is that, Prometheus scrapes metrics via HTTP in a certain format. For exposing the metrics in the Prometheus format we use what we call "exporters". These exporters read system/software values and export them via HTTP for Prometheus to scrap.

For obtaining system metrics like CPU, RAM, Temperature we use node-exporter For XMRig metrics we use xmrig-exporter

XMRig setup

Already existing XMRIG

If you are already mining using XMRig, make sure you expose the HTTP API, i.e

xmrig --http-enabled --http-host 0.0.0.0 --http-port 8080 --api-worker-id your-unique-worker-id # other params here

You'll then need to run node-exporter on your mining rig. This needs to run on your mining rig so it can obtain system stats.

xmrig-exporter does not really need to run on your miner, although it's most common to run your exporter next to your apps.

For runinng those, you can use the provided compose file or you can run them directly on the host OS, that's up to you.

XMRig on Docker

If you want to run the miner itself on Docker you can simply run the provided compose file on your mining rig. First edit the relevant parameters (i.e pool, algorithm) and just docker-compose up -d This will spin up xmrig, node-exporter and xmrig-exporter

Note that I'm using an xmrig image that I built my self, I recommend you don't trust me and build it yourself instead. You can get the Dockerfile from here (you might alsa wanna build the GCC base image too)

Monitoring setup

To run the monitoring stack, we'll use another host. You probably don't want to run Prometheus and Grafana on your mining rig. You can use an rpi, or a VM or whatever.

To run Prometheus and Grafana, first edit the /monitoring/.env file with a new user/password for Grafana.

The compose file located on /monitoring includes Prometheus, Grafana, alert-manager and nginx-proxy. You can setup alert-manager to send you hashrate alarms or temperature (rules are not currently setup, but the software is) nginx-proxy is used to access the URLs via fqdn instead of ip/port addresses. It uses .localhost as an example but you could also use your custom FQDN with your DNS server.

You'll then need to edit the /monitoring/prometheus/prometheus.yml to provide the FQDN or IP address of your mining rigs for scraping. Change the job name and the targets.

scrape_configs:
  - job_name: "my-miner-job"
    static_configs:
      - targets: ["<ip or miner fqdn>:9100"] # node-exporter
      - targets: ["<ip or miner fqdn>:9189"] # xmrig-exporter

If you docker-compose up on /monitoring you should be able to access grafana.localhost and login with your creds. It should create the Prometheus data source and dashboards by itself so it should be ready to get started.

Testing

For testing the setup, you can run the miner and monitoring on the same host on your workstation. In the scrape_configs add your workstation IP

Troubleshooting

You can access prometheus.localhost and view the targets to see if prometheus is able to reach your miners.

Also docker logs prometheus docker logs grafana

Feel free to create new issues and I'll try to help you.