Monitoring with Prometheus, Grafana, and Loki

Having a robust monitoring stack on Kubernetes is essential to ensure the reliability, performance, and security of your applications and infrastructure.With real-time dashboards and correlated insights, the teams can optimize resource usage, enforce SLOs, and minimize downtime for modern cloud-native environments.

🎞️ Watch Video

Prerequisites:

• Kubernetes cluster
• Helm 3.x 0r 4.x
• kubectl configured

Kube-prometheus-stack

Set Up Namespace and Helm Repositories

kubectl create namespace monitoring

Add the Prometheus Helm Repository

helm repo add prometheus-community https://prometheus-community.github.io/helm-charts && helm repo update

helm show values prometheus-community/kube-prometheus-stack > prometheus-stack-values.yaml

helm install prometheus prometheus-community/kube-prometheus-stack --namespace monitoring --version 79.8.2 -f prometheus-stack-values.yaml

Check the status of the deployment

kubectl get pods -n monitoring

kubectl describe pod prometheus-grafana-<pod-id> -n monitoring

kubectl logs prometheus-grafana-<pod-id> -n monitoring -c grafana

Accessing the grafana dashboard using portforward

kubectl port-forward svc/prometheus-grafana -n monitoring 3000:80

Loki

Deploying Loki for Persistent Log Aggregation

helm repo add grafana https://grafana.github.io/helm-charts
helm repo update
helm upgrade --install loki grafana/loki-stack --version 2.10.3 -f loki-value.yaml -n monitoring 

🔗 Reference Links:

• Kube prometheus stack

• GitHub-Repo