Health check with Uptime Kuma, complete setup guide in kubernetes

frustrated because didn't monitor service health I have been in this situation a lot. The services we set up went down for some reason and we are not aware of it. The customer knows about it the first and let us know. This is one of the worst situation to be in. All the services that we setup should be under our control or we should be the one to know about it or we should know about the failure as soon as things goes wrong.

The application we deploy might go wrong any time and it’s possible that we are not aware of it. To bring us all out of this worst situation we need to setup a monitoring tools that will continuously monitor our sytem or appliation and let us know as soon as something goes wrong.

how to keepup with service health (uptime)

There are multiple ways to do it:

you can write script of your own, which needs a way more time investment
install monitoring tools nagios, prometheus with alertmanager and grafana
use third party services like NewRelic, DataDog and many more in the market
setup an opensource tool like uptime kuma. The choice is yours which one you prefer. It also depends on the budget you have. Here we are going to use the opensource tool called uptime kuma which I found to be the best tool so far for the purpose.

Why uptime kuma?

I find it way easier to setup then any other tools available
The configuration part is pretty easy and has many options available.
It’s opensource.
The best part is it provides a monthly SLA report for the service being monitored.

Our Goal

We will setup uptime kuma using helm chart
we will setup a simple nginx service and try to monitor it in uptime kuma
The notification will be sent to MicroSoft teams as an alert.

Installing uptime kuma with helm chart:

Adding helm chart repo and installing it on uptime-kuma namespace

helm repo add uptime-kuma https://helm.irsigler.cloud
helm install uptime-kuma uptime-kuma/uptime-kuma -n uptime-kuma --create-namespace

Make necessary changes as per your requirements in kuberenetes values in helm chart values

These are few things you can do for setting up domain for the uptime kuma.

update helm chart to enable ingress and setup proper domain for the service
update helm chart to add proper domain to set for the ingress Here: we will simply check by port-forwarding the service

Deploy simple nginx app and monitor with uptime

Simply create a deployment and service for testing purpose. I will be creating it on “testing” namespace.

deployment.yaml

apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-deployment
  namespace: testing
  labels:
    app: nginx
spec:
  replicas: 1
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: nginx:latest
        ports:
        - containerPort: 80

service.yaml

apiVersion: v1
kind: Service
metadata:
  name: nginx-service
  namespace: testing
spec:
  selector:
    app: nginx
  ports:
    - protocol: TCP
      port: 80
      targetPort: 80
  type: ClusterIP

Finally configuring the uptime kuma setting up monitor and notification:

We have successfully deployed uptime using helm chart, now lets do necessary setup on uptime to monitor our nginx application.

add the new monitor in uptime kuma
for us as we are monitoring the kubernetes service with no external url or might not even have it in case of kubernetes so we would choose to do health check with TCP protocol set the proper names for monitor, for url for our case it will be nginx-service.testing.svc.cluster.local