dytruong / k8s-beginer Goto Github PK

• install Docker as container runtime for minikube
• install kubectl
• install minikube (optional)

physical_machine -> [k8s node] -> [pods]

• pods is smalled unit in k8s which represents a single instance. This pod can contain one container (or shared container with IP, volume, ...)
• deployment will managed pods such as scaling, versioning.
• service will manage networking and load banlancing to access pods within k8s cluter.

Create single pod in k8s

kubectl run kbnginx --image=nginx

check pods in default namespace

kubectl get pods

check pods in specific namespace

kubectl get pods --namespace=kube-system

get detail infor pod (namepace, IP, node, statuu, ... of pod)

kubectl describe pod kbnginx

ssh to minikube

minikube ssh

get pods ip and node info same as the docker ps

kubectl get pods -o wide

delete pod

kubectl delete pod kbnginx

create deployment

kubectl create deployment nginx-deployment --image=nginx

get deployments pod

kubectl get deployments

get details deployment

kubectl describe deployment nginx-deployment

scale up pods in deployment

kubectl scale deployment nginx-deployment --replicas=3

expose port of pods inside the deployment to outside

kubectl expose deployment nginx-deployment --port=8080 --target-port=80

get IP address of the deployment. Services was created with specific IP address for the deployment

kubectl get services

get detail of the service

kubectl describe service nginx-deployment

delete deployment

kubectl delete deployment nginx-deployment kubectl delete service nginx-deployment

docker build . -t harunaga113/k8s-web-hello:latest docker login docker push harunaga113/k8s-web-hello:latest

create deployment image

kubectl create deployment k8s-web-hello --image=harunaga113/k8s-web-hello

-> create service with type=NodePort

kubectl expose deployment k8s-web-hello --type=NodePort --port=3000

Or create service with type=LoadBalancer (by default, this service will get the IP address of loadbalancer and input to EXTERNAL-IP. But there are no loadbalancer, it will in pending status and work as NodePort type)

kubectl expose deployment k8s-web-hello --type=LoadBalancer --port=3000

-> start service

minikube service k8s-web-hello

After build image succeed and start deployment service done. Procceed to update the new image

kubectl set image deployment k8s-web-hello k8s-web-hello=harunaga113/k8s-web-hello:1.0.1

Roll out deployment to check status.

kubectl rollout status deploy k8s-web-hello

-> Create the deployment (deployment.yaml) create file deployment.yaml -> input file as template here - https://kubernetes.io/docs/reference/kubernetes-api/

kubectl apply -f deployment.yaml

-> Create the service (service.yaml)

kubectl apply -f service.yaml

-> Stop all

kubectl delete -f deployment.yaml -f service.yaml

In the different deployment as the same node, we can connect to other service use the service name (for ex: service name is nginx, which will be http://nginx)

Just split them by ---

Because the IP address was provided by Kubenete node and assign automatically and not stable to use.

2. How to public connection (ex: nginx) of pod inside of the k8s node?

• Create the deployment -> create the service -> public

1. How to scale and manage automatically in k8s?

• Use kubectl deployment to create pod, This pod after created will be managed by the deployment.

1. What is deployment in k8s?

• This use to manage and scale-up scale-down pods.

1. How the deployment can keep connect to pod?

• Originally, deployment is seperated with pod. This was connected through Selector connect to Labels in pod.

1. How many pods can connect to deployment?

• Many, many.

1. Why we need to load balancer for the multiple nodes in cluster?

• Because the deployment and pods can share to the different nodes in the same cluster.

• redhat/centos:

systemctl disable firewalld --now

• ubuntu:

ufw disable

curl -sfL https://get.k3s.io | INSTALL_K3S_VERSION=v1.25.11+k3s1 sh -

curl -LO https://dl.k8s.io/release/v1.25.11/bin/linux/amd64/kubectl chmod +x ./kubectl mv ./kubectl /usr/local/bin/

curl -sfL https://get.k3s.io | INSTALL_K3S_VERSION=v1.25.11+k3s1 K3S_URL=https://<IP_SERVER_CLUSTER>:6443 K3S_TOKEN=</var/lib/rancher/k3s/server/node-token> sh -

curl https://raw.githubusercontent.com/helm/helm/main/scripts/get-helm-3 | bash

helm repo add rancher-latest https://releases.rancher.com/server-charts/latest

kubectl create namespace cattle-system

kubectl apply -f https://github.com/cert-manager/cert-manager/releases/download/v1.11.0/cert-manager.crds.yaml

helm repo add jetstack https://charts.jetstack.io

helm repo update

helm install cert-manager jetstack/cert-manager
--namespace cert-manager
--create-namespace
--version v1.11.0

helm install rancher rancher-latest/rancher
--namespace cattle-system
--set hostname=<IP_SERVER_RANCHER>.sslip.io
--set replicas=1
--set bootstrapPassword=<PASSWORD_LOGIN_RANCHER>
--version 2.7.5

kubectl port-forward svc/argocd-server -n argocd 8080:443

argocd app create k8s-web --repo [email protected]:dytruong/deploy-to-k8s-labs.git --path try01 --dest-server https://kubernetes.default.svc --dest-namespace k8s-web-nginx

argocd app create authenticate-service --repo [email protected]:dytruong/deploy-to-k8s-labs.git --path authenticate-service --dest-server https://kubernetes.default.svc --dest-namespace authenticate-service

step 1: Build docker image with version tag

docker build . -t <docker_hub_username>/<docker_image>:<image_version>

step 2: Push docker image to docker hub (or others) Note: Authenticate before push

docker login

docker push <docker_hub_username>/<docker_image>:<image_version>

step 3: Init helm source git

• Create git repository to manage helm source
• Create helm charts

step 4: use argocd to deploy k8s

argocd app create <app_name> --repo <git_url> --path <folder_inside_git> --dest-server <https://kubernetes.default.svc | if argocd inside k8s cluster> --dest-namespace <namespace_app>

kubectl create namespace argocd

helm repo add dandydev https://dandydeveloper.github.io/charts helm dependency build chart helm upgrade --install --set server.ingress.enabled="true" --namespace="argocd" --values chart/values.yaml --values custom_values.yaml argocd-deploy chart

https://github.com:dytruong/k8s-beginer.git