Monitoring-Prometheus-Grafana
Deployment of Prometheus, Alert Manager, Grafana and Node Exporter on Kubernetes Cluster for SAS Viya 4 Monitoring
Install Prometheus on Kubernetes Cluster
# mkdir ~/monitoring
# git clone https://github.com/bibinwilson/kubernetes-prometheus
# cd ~/monitoring/kubernetes-prometheus
# kubectl create namespace monitoring
# kubectl create -f clusterRole.yaml
# kubectl create -f config-map.yaml
# kubectl create -f prometheus-deployment.yaml
# kubectl get pod -n monitoring
# kubectl get deployments --namespace=monitoring
# kubectl get pods --namespace=monitoring
NAME READY STATUS RESTARTS AGE
prometheus-deployment-54686956bd-hdsnx 1/1 Running 0 2m59s
# kubectl port-forward prometheus-deployment-54686956bd-hdsnx 8080:9090 -n monitoring
# kubectl create -f prometheus-service.yaml --namespace=monitoring
URL: http://k8s-master-1:30000/
http://192.168.1.1:30000
Internal Prometheus Service: http://10.244.73.86:9090
You should have a working Prometheus setup up and running
# cd ~/monitoring
# git clone https://github.com/bibinwilson/kubernetes-alert-manager.git
# cd ~/monitoring/kubernetes-alert-manager/
# kubectl create -f AlertManagerConfigmap.yaml
# kubectl create -f AlertTemplateConfigMap.yaml
# kubectl create -f Deployment.yaml
# kubectl get pods --namespace=monitoring
NAME READY STATUS RESTARTS AGE
alertmanager-8fb4ccb87-bzrwv 1/1 Running 0 39s
prometheus-deployment-54686956bd-hdsnx 1/1 Running 0 30m
URL: http://k8s-master-1:31000/#/alerts
You should have a working Prometheus setup up and running
# cd ~/monitoring
# git clone https://github.com/bibinwilson/kubernetes-grafana.git
# cd ~/monitoring/kubernetes-grafana/
# vi grafana-datasource-config.yaml
#### Change the URL to actual Prometheus URL
"url": "http://k8s-master-1:30000/"
# kubectl create -f grafana-datasource-config.yaml
# kubectl create -f deployment.yaml
# kubectl create -f service.yaml
# kubectl get pods --namespace=monitoring
NAME READY STATUS RESTARTS AGE
alertmanager-8fb4ccb87-bzrwv 1/1 Running 0 65m
grafana-548fdc7598-ngjps 1/1 Running 0 2m13s
prometheus-deployment-54686956bd-hdsnx 1/1 Running 0 95m
# kubectl get po -n monitoring
NAME READY STATUS RESTARTS AGE
alertmanager-8fb4ccb87-bzrwv 1/1 Running 0 67m
grafana-548fdc7598-ngjps 1/1 Running 0 3m37s
prometheus-deployment-54686956bd-hdsnx 1/1 Running 0 96m
URL: http://k8s-master-1:32000/ (admin/admin) (You will be promted for change password)
http://192.168.1.1:32000
Install Prometheus Node Exporter
You should have a working Prometheus setup up and running
# cd ~/monitoring
# git clone https://github.com/bibinwilson/kubernetes-node-exporter
# cd ~/monitoring/kubernetes-node-exporter
# kubectl create -f daemonset.yaml
# kubectl get daemonset -n monitoring
NAME DESIRED CURRENT READY UP-TO-DATE AVAILABLE NODE SELECTOR AGE
node-exporter 3 3 0 3 0 <none> 7s
# kubectl create -f service.yaml
# kubectl get service -n monitoring
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
alertmanager NodePort 10.99.38.176 <none> 9093:31000/TCP 83m
grafana NodePort 10.109.158.32 <none> 3000:32000/TCP 18m
node-exporter ClusterIP 10.100.152.11 <none> 9100/TCP 118s
prometheus-service NodePort 10.109.155.37 <none> 8080:30000/TCP 103m
# kubectl get endpoints -n monitoring
NAME ENDPOINTS AGE
alertmanager 10.244.73.119:9093 81m
grafana 10.244.73.73:3000 16m
node-exporter 10.244.69.47:9100,10.244.73.101:9100,10.244.73.12:9100 15s
prometheus-service 10.244.73.86:9090 101m
