K8s-cluster-setup

https://github.com/mialeevs/kubernetes_installation_docker

Step 1-7 --> execute on master and worker both
Step 8,10 and 11 --> on master
Step 9 --> on worker

create machines for master and worker (minimum 2 CPU and 2 GB are required) with ubuntu operating system
===================================================================================================
1. Add GPG key for Docker

sudo su -
wget -O - https://download.docker.com/linux/ubuntu/gpg > ./docker.key #download docker.key
gpg --no-default-keyring --keyring ./docker.gpg --import ./docker.key #import docker.gpg from docker.key
gpg --no-default-keyring --keyring ./docker.gpg --export > ./docker-archive-keyring.gpg # export docker.gpg to cache
sudo mv ./docker-archive-keyring.gpg /etc/apt/trusted.gpg.d/

2. Add docker repository and install docker

sudo add-apt-repository "deb [arch=amd64] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable"
sudo apt update
sudo apt install git wget curl -y
sudo apt install -y docker-ce
#docker and containerd get install while installing docker both are the product of same company.
#containerd is the pre-requisite for installing docker

3. Install cri-dockerd for Docker support by k8s

VER=$(curl -s https://api.github.com/repos/Mirantis/cri-dockerd/releases/latest|grep tag_name | cut -d '"' -f 4|sed 's/v//g') #get the lastest version of docker
echo $VER # check the latest docker version
wget https://github.com/Mirantis/cri-dockerd/releases/download/v${VER}/cri-dockerd-${VER}.amd64.tgz #download tarball of CRI packages of docker
tar xvf cri-dockerd-${VER}.amd64.tgz #extract the tarball
ls -al # check the present ownership of cri-dockerd
chown -R root:docker cri-dockerd # change the ownership from your user to root user
ls -al cri-dockerd #check the actual package
sudo mv cri-dockerd/cri-dockerd /usr/local/bin/ # move the CRI packages (purpose of moving to /usr/local/bin so that any user can access that , /usr/bin only root and previledge user can access)

wget https://raw.githubusercontent.com/Mirantis/cri-dockerd/master/packaging/systemd/cri-docker.service # download servic file of CRI
wget https://raw.githubusercontent.com/Mirantis/cri-dockerd/master/packaging/systemd/cri-docker.socket # download socket file of CRI
ls -al #check the service and socket files
sudo mv cri-docker.socket cri-docker.service /etc/systemd/system/ # move the both the file in system service directory
ls -al /etc/systemd/system/ # check the files in system directory

sudo sed -i -e 's,/usr/bin/cri-dockerd,/usr/local/bin/cri-dockerd,' /etc/systemd/system/cri-docker.service #Correcting the service file path (in cri-docker.service the default path is /usr/bin/cri-dockerd but our file is copied to /usr/local/bin/)
# reload then enabling the service and socket
sudo systemctl daemon-reload
sudo systemctl enable cri-docker.service
sudo systemctl enable --now cri-docker.socket

4. Add the GPG key for kubernetes

mkdir -p /etc/apt/keyrings/ #-p ==create the directory only if it not exists
curl -fsSL https://pkgs.k8s.io/core:/stable:/v1.28/deb/Release.key | sudo gpg --dearmor -o /etc/apt/keyrings/kubernetes-apt-keyring.gpg


5. Add kubernetes repository

echo "deb [signed-by=/etc/apt/keyrings/kubernetes-apt-keyring.gpg] https://pkgs.k8s.io/core:/stable:/v1.28/deb/ /" | sudo tee /etc/apt/sources.list.d/kubernetes.list  #download and export (you can change the version(v1.28) as your wish but 2 less version considere as more stable)


6. Update the repositories

sudo apt-get update


7. Install Kubernetes packages.

sudo apt-get install -y kubelet kubeadm kubectl


8. Execute below command on master node to initialize kubernetes cluster

sudo kubeadm init --apiserver-advertise-address=<master-ip-address> --cri-socket unix:///var/run/cri-dockerd.sock  --pod-network-cidr=10.244.0.0/16
# replace the <master-ip-address> with server private ip (cmd- ip a) this argument is not mandatory but if your machine have multiple ip then you have to mention one
#--cri-socket unix:///var/run/cri-dockerd.sock -- this is also optional k8s can auto detect the CRI-socket but if you are useing 2 different CRI (docker && CRI-O) then you have to mention one
#--pod-network-cidr=10.244.0.0/16 --this is the private network , here we are mentioning the subnet ip (i.e 10.244.0.0/16)

output of init command will give you kubeadm join command that needs to be executed on worker node
in case join command is lost we can get it again
kubeadm token create --print-join-command

#from the o/p file , kuectl need to know the address of master so below config is helps kubectl to know the master from .kube/config and get authorization to login in master
  mkdir -p $HOME/.kube
  sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
  sudo chown $(id -u):$(id -g) $HOME/.kube/config



9. Add the worker node, execute below command on worker node

kubeadm join <master-ip-address>:6443 --cri-socket unix:///var/run/cri-dockerd.sock --token st99ja.3bulo53e3wlmunbk --discovery-token-ca-cert-hash sha256:41dfafa343841da6a5460f8fa89647192ec1c7f46b3029c280ff4063b940041f

#replace this command with the o/p of step number 8 , just add --cri-socket unix:///var/run/cri-dockerd.sock to determine the CRI for example 
kubeadm join 10.128.0.2:6443 --cri-socket unix:///var/run/cri-dockerd.sock --token zfh90s.pkz92x7k9f9jw58d \
        --discovery-token-ca-cert-hash sha256:1651a687c336501f371170c078d160b7c72b067baf749dc4955ab88828f98b7c
		
# to run cmd in Worker nodes--> kubectl get nodes (you have to copy .kube/config file to workernode from master


10. Deploy pod network on master, network will be created automatically on all nodes that are part of cluster

There are many network plugins available calico and flannel are 2 most popular ones

Flannel -->
kubectl apply -f https://github.com/coreos/flannel/raw/master/Documentation/kube-flannel.yml

Calico --> 

kubectl create -f https://raw.githubusercontent.com/projectcalico/calico/v3.25.0/manifests/tigera-operator.yaml
curl https://raw.githubusercontent.com/projectcalico/calico/v3.25.0/manifests/custom-resources.yaml -O
kubectl create -f custom-resources.yaml


11. Test operations

kubectl get nodes
kubectl run firstpod --image=nginx
kubectl expose po firstpod --type=NodePort --port 80 --name firstsvc