Deploy Demo Application

Now that Prometheus is monitoring the Kubernetes infrastructure and cluster endpoints are reachable, this lesson shows how to instrument and deploy a simple application so Prometheus (installed via the Prometheus Operator) can scrape its metrics. We’ll deploy a Node.js Express demo app that exposes a few HTTP endpoints and Prometheus-style metrics via the swagger-stats middleware. The app listens on port 3000 and serves metrics at /swagger-stats/metrics — the path Prometheus should scrape.

Prometheus scrapes an HTTP metrics endpoint. In this example the application exposes metrics at /swagger-stats/metrics. Configure Prometheus (or create a ServiceMonitor) to scrape that path on the Service that fronts this application.

Application code

Below is the Node.js application (index.js). It uses swagger-stats to expose Prometheus-compatible metrics and provides a few simple endpoints you can test.

const express = require("express");
const swStats = require("swagger-stats");
const app = express();

// Expose Prometheus-style metrics at /swagger-stats/metrics
app.use(swStats.getMiddleware());

app.get("/", (req, res) => {
  res.send("Hello World!");
});

app.get("/comments", (req, res) => {
  res.send("Comments");
});

app.get("/threads", (req, res) => {
  res.send("Threads");
});

app.get("/replies", (req, res) => {
  res.send("replies");
});

const PORT = process.env.PORT || 3000;
app.listen(PORT, () => {
  console.log(`API listening on port ${PORT}`);
});

Key points:

The app serves simple responses at /, /comments, /threads, and /replies.
Prometheus-style metrics are exposed at /swagger-stats/metrics.
The app listens on port 3000; ensure the container and Service route traffic to this container port.

Quick reference: endpoints and their purpose

Endpoint	Purpose
`/`	Basic health / greeting endpoint
`/comments`	Example API endpoint
`/threads`	Example API endpoint
`/replies`	Example API endpoint
`/swagger-stats/metrics`	Prometheus metrics endpoint (scrape target)

Dockerfile

This Dockerfile builds the application image and exposes port 3000 to match the app.

FROM node:16

# Create app directory
WORKDIR /usr/src/app

COPY package*.json ./

RUN npm install
# If you are building your code for production
COPY . .

EXPOSE 3000

CMD [ "node", "index.js" ]

Build and push the image to your container registry (replace with your own image name/repository):

sudo docker build -t sanjeevkt720/prometheus-demo:latest .
sudo docker push sanjeevkt720/prometheus-demo:latest

Example push confirmation:

The push refers to repository [docker.io/sanjeevkt720/prometheus-demo]
latest: digest: sha256:c6d9b4622d57f5ea5891226c44e0a16a928b60eefdcb9ee29b2109159e35992c size: 1234

(Your digest and size will differ.)

Kubernetes manifests

Create a single manifest file api-deploy.yaml that includes both the Deployment and the Service. The Deployment runs two replicas of the container and exposes container port 3000. The Service is a ClusterIP Service that forwards port 3000 to pods labeled app: api. api-deploy.yaml:

apiVersion: apps/v1
kind: Deployment
metadata:
  name: api-deployment
  labels:
    app: api
spec:
  replicas: 2
  selector:
    matchLabels:
      app: api
  template:
    metadata:
      labels:
        app: api
    spec:
      containers:
        - name: api
          image: sanjeevkt720/prometheus-demo:latest
          ports:
            - containerPort: 3000
---
apiVersion: v1
kind: Service
metadata:
  name: api-service
  labels:
    job: node-api
    app: api
spec:
  type: ClusterIP
  selector:
    app: api
  ports:
    - name: web
      protocol: TCP
      port: 3000
      targetPort: 3000

Notes:

The Deployment’s selector.matchLabels: app: api must match the Pod template labels.
The Service selector: app: api must match the Pod labels so traffic reaches the pods.
ClusterIP keeps the Service internal to the cluster. To expose externally, change type to NodePort or LoadBalancer depending on your environment.

Ensure the container’s exposed port matches the Service targetPort. In this example both are 3000. Mismatched ports commonly cause “connection refused” errors.

Deploy to the cluster

Apply the manifest:

kubectl apply -f api-deploy.yaml

Expected output:

deployment.apps/api-deployment created
service/api-service created

Verify the objects:

kubectl get service api-service
kubectl get deployment api-deployment

Example outputs (values will vary):

NAME          TYPE        CLUSTER-IP      PORT(S)
api-service   ClusterIP   10.100.40.38    3000/TCP

NAME             READY   UP-TO-DATE   AVAILABLE   AGE
api-deployment   2/2     2            2           1m

Once the Deployment is ready and the Service exists, the application is reachable from inside the cluster:

Service URL (cluster-internal): http://api-service:3000/swagger-stats/metrics
Pod URL (for a specific pod): http://<pod-ip>:3000/swagger-stats/metrics

Prometheus (or the Prometheus Operator via a ServiceMonitor) should be configured to scrape http://api-service:3000/swagger-stats/metrics. To complete monitoring, create or update your Prometheus scrape configuration (or a ServiceMonitor) so Prometheus discovers and scrapes this Service. If using the Prometheus Operator, a ServiceMonitor targeting api-service with the path /swagger-stats/metrics and port web (or 3000) is the recommended approach.

Links and references

Prometheus Operator (kube-prometheus-stack): https://github.com/prometheus-operator/kube-prometheus
Prometheus: https://prometheus.io/
Kubernetes documentation: https://kubernetes.io/docs/
Docker Hub: https://hub.docker.com/

Introduction

Observability Fundamentals

Prometheus Fundamentals

PromQL

Dashboarding & Visualization

Application Instrumentation

Service Discovery

Push Gateway

Alerting

Monitoring Kubernetes

Conclusion

Scaling & Long Term Storage

Deploy Demo Application

Application code

Dockerfile

Kubernetes manifests

Deploy to the cluster

Links and references

Watch Video

​Application code

​Dockerfile

​Kubernetes manifests

​Deploy to the cluster

​Links and references

Watch Video

Application code

Dockerfile

Kubernetes manifests

Deploy to the cluster

Links and references