Skip to content

Latest commit

 

History

History
36 lines (23 loc) · 1.49 KB

monitoring.md

File metadata and controls

36 lines (23 loc) · 1.49 KB
Raw

Monitoring Torus

1) Run torusd with a monitor port

torusd supports listening for HTTP requests on a monitoring port. Running with the options:

--host $IP --port 4321

Enables this functionality. When running inside a container, this is automatically done in the entrypoint script.

2) Set up Prometheus to monitor your cluster

If you already have a Prometheus monitoring system set up, you're ready. Add these hosts and ports to be scraped.

If not, Prometheus is a fantastic monitoring tool, and Torus exports all its metrics through the monitor port under the expected /metrics path.

Getting started with Prometheus is well-documented. Adding an entry under scrape_configs for your prometheus.yaml proceeds as normal:

scrape_configs:
  # The job name is added as a label `job=<job_name>` to any timeseries scraped from this config.
  - job_name: 'torus'

    # Override the global default and scrape targets from this job every 5 seconds.
    scrape_interval: 15s
    scrape_timeout: 30s

    target_groups:
      - targets: ['localhost:4321', 'localhost:4322', 'localhost:4323', 'localhost:4324']

3) Using grafana

If you're also using grafana to build dashboards on your Prometheus metrics, then you can import the default torus dashboard from the repository or release; it lives in contrib/grafana , and customize to fit your use cases.