This Repo is for the demo for running Chaos experiment on Civo Kubernetes.
Create a K3s cluster (In this Demo we are using Civo Kubernetes cluster). You can also create K3s cluster on Katakoda ubuntu playgound or any other VM instance.
Steps:
- Signup for Civo (https://www.civo.com/signup)
- Download the Civo cli (https://github.com/civo/cli)
- Configure the cli with api key (https://github.com/civo/cli#api-keys)
- create cluster using below command
civo kubernetes create chaos - Install Litmus 2.0
kubectl create ns litmus
kubectl apply -f https://litmuschaos.github.io/litmus/2.0.0-Beta/litmus-2.0.0-Beta.yaml
kubectl get pods -n litmus
NAME READY STATUS RESTARTS AGE
litmuschaos-litmus-2-0-0-beta-frontend-6bbdb89479-lvftk 1/1 Running 0 24h
litmuschaos-litmus-2-0-0-beta-server-5fdf755679-qnvzj 2/2 Running 0 24h
litmuschaos-litmus-2-0-0-beta-mongo-0 1/1 Running 0 24h
- Deploy the application
kubectl apply -f deploy/hello.yaml
kubectl get pods
NAME READY STATUS RESTARTS AGE
helloservice-579fdcf676-2c4f4 1/1 Running 0 10m
- Eanble Gitops in Litmus
- Deploy Prometheus, Blackbox exporter and Grafana
kubectl apply -f monitoring/prometheus
kubectl apply -f monitoring/blackbox-exporter
kubectl apply -f monitoring/grafana
-
Create Grafana Dashboard by adding Prometheus as the data source and Json data from monitoring/grafana/dashboard.json
-
Create a custom pod networkloss Workflow and set monitoring to true
-
Bootstrap Flux and use the path as deploy folder to put the yaml for deployment with new image. It is using the Jinja template in the tmpl folder
flux bootstrap github --components-extra=image-reflector-controller,image-automation-controller --owner=saiyam1814 --repository=KUBECON2021EU --branch=main --path=deploy --token-auth --personal
- Setup Github Actions for build and push image and create manifest in the deploy folder