Kubernetes installation models:
-------------------------------------
1.Kubeadm
2.Kops
3.Minikube
and we do have other models from AWS(EKS),GCP(GKE),AZURE(AKS)
1. Kubernetes Cluster setup with Kubeadm on GCP
Kubeadm can be installed in Bare-metal or any vmware or any Cloud computing technologies such as GCP,AWS,AZURE.
-------------------------------------
1.Kubeadm
2.Kops
3.Minikube
and we do have other models from AWS(EKS),GCP(GKE),AZURE(AKS)
1. Kubernetes Cluster setup with Kubeadm on GCP
Kubeadm can be installed in Bare-metal or any vmware or any Cloud computing technologies such as GCP,AWS,AZURE.
in general Master cluster setup will be minimum 3 master nodes and multiple worker nodes in production grade setup.
each Master nodes have at least 4 GB RAM .and 2 CPU cores.and worker nodes should be 2GB RAM and 1 CPU core to practice the Kubernetes components.
firstly will go with Single master and 2 workers kubernetes setup in GCP.
for this we need GCP account and need to be subscribe a project. as i have done with basic steps.
creating 4 compute engines, 1 for Master , 2 for worker and 1 for kubernetes management server and this will help us to install the kubernetes libraries. OS selection is centos 7.
followed Official document and add my best practice to setup the cluster.
login to all the 4 servers and with root and follow the steps,
Pre-requisites:
execute below commands one by one to configure the Kubernetes setup. in the 3 server(master-01,worker-01 & worker-02)
1. add below entries on the 3 server under file path vi /etc/hosts file
master-01 xx.xxx.x.xx
worker-01 xx.xxx.x.xx
worker-02 xx.xxx.x.xx
2. To Stop Firewall and NetworkManager :
2. To Stop Firewall and NetworkManager :
systemctl stop firewalld NetworkManager && systemctl disable firewalld NetworkManager
3. To Disable selinux :
Go to path and edit as follow vi /etc/sysconfig/selinux
change the parameter from SELINUX=enforcing to SELINUX=disabled and save
4. To off the SWAP :
change the parameter from SELINUX=enforcing to SELINUX=disabled and save
4. To off the SWAP :
Execute the command swapoff -a
5. To disable ipv6:
5. To disable ipv6:
go the file path vi /etc/sysctl.conf and add the following entries
net.ipv6.conf.all.disable_ipv6 = 1
net.ipv6.conf.default.disable_ipv6 = 1
net.ipv4.ip_forward=1
and save the file and execute below command to check the ipv6 and ipv4 disabled.
net.ipv6.conf.all.disable_ipv6 = 1
net.ipv6.conf.default.disable_ipv6 = 1
net.ipv4.ip_forward=1
and save the file and execute below command to check the ipv6 and ipv4 disabled.
to check the status of the IPV6 and IPV4 with command sysctl -p
with all above steps we have completed with Prerequisites.
with all above steps we have completed with Prerequisites.
now will go with installation of CRI (Container Runtime environment) , in my case i used Docker.
Installation steps to Docker CE :
Installation steps to Docker CE :
execute the below steps as it is in the centos servers.
yum install yum-utils device-mapper-persistent-data lvm2
yum-config-manager --add-repo https://download.docker.com/linux/centos/docker-ce.repo
yum install docker-ce-18.06.2.ce
mkdir -p /etc/docker
Setup daemon.
cat > /etc/docker/daemon.json <<EOF
{
"exec-opts": ["native.cgroupdriver=systemd"],
"log-driver": "json-file",
"log-opts": {
"max-size": "100m"
},
"storage-driver": "overlay2",
"storage-opts": [
"overlay2.override_kernel_check=true"
]
}
EOF
To enable the docker service.
mkdir -p /etc/systemd/system/docker.service.d
Restart Docker & enable docker
systemctl daemon-reload
systemctl restart docker
systemctl enable docker
Now we have to install the Kubernetes packages , steps are as follows.
cat <<EOF > /etc/yum.repos.d/kubernetes.repo
[kubernetes]
name=Kubernetes
baseurl=https://packages.cloud.google.com/yum/repos/kubernetes-el7-\$basearch
enabled=1
gpgcheck=1
repo_gpgcheck=1
gpgkey=https://packages.cloud.google.com/yum/doc/yum-key.gpg https://packages.cloud.google.com/yum/doc/rpm-package-key.gpg
exclude=kubelet kubeadm kubectl
EOF
to specific version selection just run the command with any one of the package: yum search kubelet --show-duplicate --disableexcludes=kubernetes
now im going to select kubelet-1.18.2-0.x86_64 version packages,
yum install -y kubelet-1.18.2-0.x86_64 kubeadm-1.18.2-0.x86_64 --disableexcludes=kubernetes
yum install -y kubectl-1.18.2-0.x86_64 --disableexcludes=kubernetes
# to enable the kubelet
systemctl enable --now kubelet
As a requirement for Linux Node's iptables to correctly see bridged traffic, we should ensure
net.bridge.bridge-nf-call-iptables is set to 1 in our sysctl configcat <<EOF > /etc/sysctl.d/k8s.conf
net.bridge.bridge-nf-call-ip6tables = 1
net.bridge.bridge-nf-call-iptables = 1
EOF
sysctl --system
go to master-01 with root user
execute the below commands
kubeadm config images pull
in my case i'm not using any loadbalencers like ha-proxy, so just simply executing the kubeadm init.
kubeadm init
you can see that below message :
Your Kubernetes control-plane has initialised successfully!
To start using your cluster, you need to run the following commands.
mkdir -p $HOME/.kube
sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
sudo chown $(id -u):$(id -g) $HOME/.kube/config
Then go to each worker node and execute the below Join command join as root: ( this will get after initiate the kubeadm init
kubeadm join 10.128.15.227:6443 --token 8ynfsy.t2gfni04ilrdzuy1 \
--discovery-token-ca-cert-hash sha256:f25b63ea2956d1fb3d2fd1938f8a8217a4496b2a4071e2ca7d12e7fbcb70b517
on worker-01
on Worker-02
so we are done with the cluster set up to check the status of the cluster.
go to management VM instance and install the kubectl then add the .kube file and execute the below command for that
copy the content from cat /root/.kube/config on master-01 and paste it in to the management server path vi /root/.kube/config
kubectl get nodes
but it is not desired output of the clusters nodes. please do the below steps to fix those.
to set all the worker and master in the same network, execute below command from management server to get the Status of the nodes.
kubectl apply -f "https://cloud.weave.works/k8s/net?k8s-version=$(kubectl version | base64 | tr -d '\n')"
to set the Role of the workers ,execute the be command from management servers
kubectl label node worker-01 node-role.kubernetes.io/worker=worker
kubectl label node worker-02 node-role.kubernetes.io/worker=worker
now all the issues got fixed and we got the desired output of the Nodes.
CRI:
https://kubernetes.io/docs/setup/production-environment/container-runtimes/
Kubernetes configurations with kubeadm:
https://kubernetes.io/docs/setup/production-environment/tools/kubeadm/create-cluster-kubeadm/
Kubectl:
https://kubernetes.io/docs/reference/kubectl/overview/