Unofficial Prometheus Exporter for Rancher

Note : This project is not officially supported by Rancher/Suse.

Quickstart

1. Enable monitoring in the Rancher Management Cluster, aka local cluster

2. Apply the manifest from this repo : kubectl apply -f ./manifests/exporter.yaml

3. You can set timers by environment variable:

   TIMER_TICKER=10 - how ofter read data, in seconds

   TIMER_GET_LATEST_RANCHER_VERSION=1 - how often get latest rancher version from github, in minutes

Grafana Dashboards

./manifests/grafana-dashboard includes a basic dashboard in JSON format that can be imported into Grafana.
./manifests/grafana-dashboard-projects.json includes a Rancher project-focused dashboard in JSON format that can be imported into Grafana.
./manifests/grafana-dashboard-all-cr.json includes a dynamic dashboard showing counts for all Rancher custom resources (*.cattle.io).
./manifests/grafana-dashboard-nodes.json includes a dynamic dashboard showing global node metadata across all managed clusters (ie Kernel/OS versions, roles, etc)
./manifests/grafana-dashboard-backups-json includes a dynamic dashboard showing details about Rancher backup operator jobs.

Example Screenshots

Developing

By default, the exporter will use in-cluster authentication via a associated service account

External cluster config

To test using external authentication via the local kubeconfig, set the following environment variable:

export RANCHER_EXPORTER_EXTERNAL_AUTH=TRUE

• go run main.go
• Navigate to http://localhost:8080/metrics