This document describes how to deploy RealOpInsight on a Kubernetes cluster.
The deployment can be achieved using Helm or kubectl + Kustomize. Select the appropriate section below to getting started.
First review the default configuration settings in the deployment ConfigMap: kustomize/realopinsight-config.yaml.
The manifests are configured to achieve the deployment in the namespace monitoring. The namespace is assumed to exist, otherwise create it first.
Use the below command to trigger the deployment.
kubectl -n monitoring apply -k ./manifests/kustomize
# Open http://localhost:4583/ui in your browser (default credentials : admin/password).Assuming you have a Linux terminal with Helm3 installed and able to access a KUBECONFIG file to deploy resources on Kubernetes, the following command shall install an instance in the monitoring namespace. The namespace must exist, it can be changed to another value.
helm upgrade --namespace monitoring --install realopinsight manifests/helm/realopinsight/On success, the deployment does create a Kubernetes ClusterIP service named realopinsight.
This service exposes the RealOpInsight's web interface (port 80) as well as its Prometheus exporter (port 4584).
Their in-cluster URLs are the following:
- Web UI:
http://realopinsight.monitoring/ui. - Prometheus metrics:
http://realopinsight.monitoring:4583/metrics.
As a developer or for debugging purposes, you can enable port forwarding for local access as follows.
kubectl port-forward --namespace monitoring service/realopinsight 4583:80For production access, typically for operations monitoring, you must have to consider the Ingress policies defined in your organization. Contact your IT staff for additional information.