Deploying Prometheus and Grafana on top of Kubernetes
Integrating monitoring tools in the Development Pipelines
After a month of working with Docker, Kubernetes, GitHub and Jenkins, to create development pipelines, now is its the time to integrate monitoring tools into the existing pipelines, so that we can even automate the monitoring actions on the cloud.
For performing the monitoring actions, we have 2 very efficient tools which work hand-in-hand, Prometheus and Grafana. In this article, we are going to implement these tools on top of Kubernetes.
The integration would be done in the following manner:
- Deploying them as pods on top of Kubernetes by creating resources Deployment, Replica Set, Pods or Services.
- Making their data to remain persistent.
- Exposing both to the outside world
First, let’s understand the working of the monitoring tools we are using:
Prometheus is an open-source system monitoring and alerting toolkit originally built at SoundCloud. Since its inception in 2012, many companies and organizations have adopted Prometheus, and the project has a very active developer and user community. It is now a standalone open source project and maintained independently of any company.
Grafana is an open-source, general-purpose dashboard and graph composer, which runs as a web application. Grafana allows you to query, visualize, alert on and understand your metrics no matter where they are stored. Create, explore, and share dashboards with your team and foster a data-driven culture.
First of all, we are assuming that Kubectl and Minikube are installed in the system.
1. Setting up the PVC for Prometheus
Prometheus by default has ephemeral storage. Therefore we need to use a Persistent Volume Storage so that data stored does not get lost in the running pod fails. For that, we are using the Kubernetes Persistent Volume Claim (PVC)
To set up we need to create the following YAML file.
2. Deploying Prometheus Docker Image
After setting up the PVC, we now have to install Prometheus. For that first, we need to create a service to export the Prometheus using NodePort. After that, we can use the Prometheus Docker Image on Dockerhub to install the tool in Minikube.
The final YAML file would look like this:
3. Setting up the PVC for Grafana
(Same codes with minor changes)
4. Deploying Grafana Docker Image
(Same codes with minor changes)
5. Creating the Kustomization file
In this file, we will be mentioning the sequence in which all the files need to be created. After running this file, it will do everything, Prometheus and Grafana will have been deployed on Kubernetes and have been connected to PVC for no data loss. Both of the pods will be exposed to the outside world. Using Kubernetes, we don’t need to worry about crashes, because that is the biggest advantage of using Kubernetes, and as we are using PVC, so there will be no data loss whatsoever.
To run the entire infrastructure we created, just the kustomization file has to be run with the following command:
kubectl apply -k .
6. Exposing both to the outside world
We need to open the Prometheus & Grafana in the browser using the Minikube IP to get access to the monitoring tools.
And we reached the end of the solution!!!
You can visit the repository from below: