Kismatic prometheus operator

This project contains a deployment bundle to install prometheus operator to a kubernetes cluster, provisioned by kismatic enterprise toolkit. It's based on the contrib scripts for cluster monitoring from the prometheus operator.

why not using the original contrib scripts from prometheus operator project?

• the current version (v1.6.1) has an issue in the scraping progress
• the controller manager and scheduler are not properly set up, so alerts are comming

What's in the box?

• a prometheus operator, which automatically can spawn new prometheus instances
• new TRPs: Prometheus, ServiceMonitor, AlertMananger, see documentation and GitHub Project for further info
• a monitoring interface including node exporters, kube-state-metrics and prometheus for cluster-monitoring
• a prometheus instance on http://<NODE_IP>:30900 for cluster monitoring
• a grafana instance on http://<NODE_IP>:30902 (user: admin, password: admin
• a alert manager instance on http://<NODE_IP>:30903

Requirements

• existence of kubectl
• cluster context set up for your KET cluster

Install & Teardown

Run

$ ./deploy

to install all what's in the box to the namespace "monitoring", and

$ ./teardown

to remove the workloads