https://github.com/bedus-creation/k8-learning
https://github.com/bedus-creation/k8-learning
Last synced: about 1 year ago
JSON representation
- Host: GitHub
- URL: https://github.com/bedus-creation/k8-learning
- Owner: bedus-creation
- Created: 2024-07-21T07:13:35.000Z (almost 2 years ago)
- Default Branch: main
- Last Pushed: 2025-02-03T20:42:21.000Z (over 1 year ago)
- Last Synced: 2025-02-03T21:30:04.851Z (over 1 year ago)
- Language: JavaScript
- Size: 196 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: readme.md
Awesome Lists containing this project
README
# Kubernetes Learning
* [x] Hello world application in Kubernetes
* [x] Build a custom docker image for PHP, nginx application & publish to the docker Hub
* [x] Write a deployment script for PHP,Nginx & Mysql using the custom docker image build previously
* [x] Install laravel application and configured to run inside kubernetes clusters
* [x] Scale the Laravel application horizontally
* [ ] Write more advance scaling logic to scale application based on cpu use etc.
* [ ] Load Test
* [ ] Scale MySQL database using [Vitess](https://planetscale.com/blog/what-is-vitess)
* [x] Setup Vitest
* [ ] Sharding
* [x] Scale MySQL database using [Percona](https://docs.percona.com/percona-operator-for-mysql/ps/kubernetes.html)
* [x] [setup Percona](https://github.com/bedus-creation/k8-learning/tree/main/k8s/percona#setup)
* [x] [Horizontal Scaling](https://github.com/bedus-creation/k8-learning/tree/main/k8s/percona#horizontal-scaling)
* [ ] use redis as session storage and scale it
## Pod, Service & Deployment
A Pod is the smallest deployable unit in Kubernetes. It represents a single instance of a running process in your cluster.
### Service
A Service is an abstraction that defines a logical set of Pods and provides a stable endpoint to access them. It enables communication between Pods or external clients and Pods.
Key Characteristics:
* Stable Networking: Provides a consistent IP address and DNS name, even as Pods are replaced or scaled.
* Load Balancing: Distributes traffic across multiple Pods matching its selector.
* Service Types:
* ClusterIP (default): Internal communication within the cluster.
* NodePort: Exposes the service on a port of each node for external access.
* LoadBalancer: Exposes the service externally using a cloud provider's load balancer.
* ExternalName: Maps the service to an external DNS name
### Deployment
A Deployment is a controller that manages Pods and ensures they are running in the desired state. It allows you to define and manage the number of replicas (instances), updates, and rollbacks for your application.
### Install kubectl
The Kubernetes command-line tool, kubectl, allows you to run commands against Kubernetes clusters. You can use kubectl to deploy applications, inspect and manage cluster resources, and view logs.
[Link](https://kubernetes.io/docs/tasks/tools/)
## Hello World in Kubernetes
We can start create a very first deployment using a public docker image.
```shell
kubectl create deployment nginx-deployment --image=nginx
```
The above deployment can be exposed outside the container & kubernetes using expose command.
```shell
kubectl expose deployment nginx-deploment --type=LoadBalancer --port=8080 --target-port=80
```
The above command creates a service for the 'nginx-deployment' deployment with a LoadBalancer type, enabling access to our Kubernetes cluster from outside the container.
You can use following commands to get details about pods:
| Commands | Descriptions |
|-----------------------------------------------|--------------------------------|
| `kubectl get pods` | Get all the pods |
| `kubectl get pods -o wide` | get all pods with IP addresses |
| `kubectl describe pod nignx-65cfcbff97-b8ccx` | Get the details of a pod |
You can use following commands to get details about deployments:
| Commands | Descriptions |
|------------------------------------|----------------------------|
| `kubectl get deployments` | Get all the deployments |
| `kubectl delete deployments --all` | delete all the deployments |
## Services
| Commands | Descriptions |
|--------------------------------------------|----------------------|
| `kubectl get services` | Get all the services |
| `kubectl describe service nginx-deployment` | Detail of a service |
The `kubectl get services` returns the following:
```shell
kubectl get services
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
php ClusterIP 192.168.194.147 9000/TCP 35h
kubernetes ClusterIP 192.168.194.129 443/TCP 35h
nginx NodePort 192.168.194.181 8080:32539/TCP 35h
nginx-deploment-1 LoadBalancer 192.168.194.248 198.19.249.2 8081:32479/TCP 34h
mysql NodePort 192.168.194.218 4406:32740/TCP 105m
```
* nginx-deployment-1 Service:
External IP: 198.19.249.2 (assigned via LoadBalancer) and Exposed Port (NodePort): 32479, so it can be accessed from `http://198.19.249.2:32539`
* NodePort Service (nginx):
For NodePort services, the application is accessible via: `http://:32539` Here, refers to the IP of the node running the cluster. If it's running on orbstack: `kubectl get nodes
-o wide` to get ``.
### Load-balancer
### Minikube Dashboard
| Commands | Descriptions |
|---------------------|------------------------------|
| `minikube start` | |
| `minikube dashboard | Start the minikube dashboard |
### Exposing Application
If your service type is ClusterIP (default), it’s only accessible within the cluster. Access via kubectl port-forward: Forward the service port to your local machine:
```bash
kubectl port-forward svc/nginx 8080:80
```
To check if nginx is working
```shell
kubectl logs
```
### Scaling
Scaling a deployment by running the following commands:
```shell
kubectl scale deployment --replicas=1
```
### Applying changes
```shell
kubectl apply -f k8s/nginx-service.yaml
```
## Managing .ENV
Kubernetes allows to create a secret based on .env file, and these secret can be referenced to create a deployment scripts.
```shell
kubectl create secret generic env-secret --from-env-file=./application/.env --dry-run=client -o yaml > k8s/env-secret.yaml
```
## Debugging
You can use the following command to view the details of all Pods, or specify a Pod name to inspect a particular Pod.
| Commands | Descriptions |
|-------------------------------------------------|------------------------------|
| `kubectl describe pod` | |
| `kubectl describe pod -n ` | Start the minikube dashboard |
## SSH Connection
A pod can be accessed with SSH Connection as:
| Commands | Descriptions |
|-------------------------------------------------|------------------------------|
| `kubectl exec -it -- ` | |
Example of database connection from PHP cluster:
```php
connect_error) {
die("Connection failed: ".$conn->connect_error);
}
echo "Connected successfully!";
```
## Deploying custom docker Image with k8:
Since k8 can't use local docker image, so we need to push it to the container registry. To build php image
```bash
docker build -t 9813276057/application-php:v0.0.1 ./php/
```
To build the nginx Image
```bash
docker build -t 9813276057/application-nginx ./nginx/
```
After building the images, push it to the container hub:
```bash
docker push 9813276057/application-php:v0.0.1
```
```bash
docker push 9813276057/application-nginx
```
### Vitess
```bash
DB_CONNECTION=mysql
DB_HOST=example-vtgate-ae7df4b6
DB_PORT=3306
DB_DATABASE=commerce
DB_USERNAME=user
DB_PASSWORD=
```