Skip to content

Latest commit

 

History

History
272 lines (205 loc) · 8.68 KB

readme.md

File metadata and controls

272 lines (205 loc) · 8.68 KB
Raw

k8s tutorial

Step 1. Write cloud-init manifest

Sample cloud-init file: https://cloudinit.readthedocs.io/en/latest/reference/examples.html#including-users-and-groups

Step 2. Init multipass nodes (3 master-nodes with 2G RAM and 3 work-nodes with 1.5G RAM)

multipass launch --name k8s-master{Node.ID} --memory 2G --cloud-init cloud-init.yml --disk=15G
multipass launch --name k8s-node{Node.ID} --memory 1536M --cloud-init cloud-init.yml --disk=15G

Step 3. Make ansible inventory file with nodes for test successfully connection

[k8s-master]
k8s-master1 ansible_host=10.57.172.124
k8s-master2 ansible_host=10.57.172.117
k8s-master3 ansible_host=10.57.172.216

[k8s-worker]
k8s-worker1 ansible_host=10.57.172.213
k8s-worker2 ansible_host=10.57.172.221
k8s-worker3 ansible_host=10.57.172.145

[k8s-cluster:children]
k8s-master
k8s-worker

And test connection with default ansible ping command

ansible --inventory ~/cloud-init/ansible/inventory/cluster.ini k8s-cluster -m ping

Step 3. Clone git kubespray repo and install all requirements

Clone repo

git clone git@github.com:kubernetes-sigs/kubespray.git

Install python requirements

pip install -r requirements.txt

Make copy of dir

cp -R ./kubespray/inventory/sample ./kubespray/inventory/cluster

Edit cluster nodes for ansible

[all]
master1 ansible_host=10.57.172.124
master2 ansible_host=10.57.172.117
master3 ansible_host=10.57.172.216
worker1 ansible_host=10.57.172.213
worker2 ansible_host=10.57.172.221
worker3 ansible_host=10.57.172.145
		
# ## configure a bastion host if your nodes are not directly reachable
# [bastion]
# bastion ansible_host=x.x.x.x ansible_user=some_user

[kube_control_plane:children]
kube_master

[kube_master]
master1
master2
master3

[etcd:children]
kube_master

[kube_node]
worker1
worker2
worker3

[calico_rr]

[k8s_cluster:children]
kube_master
kube_node
calico_rr

Edit cluster configuration in group_vars/k8s_cluster/k8s-cluster.yaml (For expert)

Run ansible-playbook installation tool (Need -b or --become)

ansible-playbook --inventory=./inventory/cluster/inventory.ini ./cluster.yml --become

Now we have installed cluster on machines. We should test them

For test cluster, connect to master-node and test cluster:

kubectl cluster-info

If we have error by 8080 port, move kube config only on master-nodes

mkdir -p $HOME/.kube
sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
sudo chown $(id -u):$(id -g) $HOME/.kube/config

Install kubectl on local workstation for connecting to cluster

# download file
curl -LO https://storage.googleapis.com/kubernetes-release/release/`curl -s https://storage.googleapis.com/kubernetes-release/release/stable.txt`/bin/linux/amd64/kubectl

# set as executable
chmod +x ./kubectl

# move to /usr/local/bin folder for access from any place in terminal
sudo mv ./kubectl /usr/local/bin/kubectl

# test them
kubectl version --client

For establish connection from local kubectl to k8s master node we can get configuration file from master-node

# download file from remote host
scp 10.57.172.124:/etc/kubernetes/admin.conf ~/cloud-init/admin.conf

# update ENV variable

echo "export KUBECONFIG=~/cloud-init/admin.conf" >> ~/.bash_profile

# apply changes
source ~/.bash_profile

Now, we need test connection to cluster;

kubectl version

If we see message "connection refused", change IP-address from localhost in config-file to master-node address and test connection

Get k8s nodes

kubectl get nodes -o wide

NAME      STATUS   ROLES           AGE    VERSION   INTERNAL-IP     EXTERNAL-IP   OS-IMAGE             KERNEL-VERSION      CONTAINER-RUNTIME
master1   Ready    control-plane   10h    v1.25.6   10.57.172.124   <none>        Ubuntu 22.04.1 LTS   5.15.0-58-generic   containerd://1.6.15
master2   Ready    control-plane   144m   v1.25.6   10.57.172.117   <none>        Ubuntu 22.04.1 LTS   5.15.0-58-generic   containerd://1.6.15
master3   Ready    control-plane   143m   v1.25.6   10.57.172.216   <none>        Ubuntu 22.04.1 LTS   5.15.0-58-generic   containerd://1.6.15
worker1   Ready    <none>          142m   v1.25.6   10.57.172.213   <none>        Ubuntu 22.04.1 LTS   5.15.0-58-generic   containerd://1.6.15
worker2   Ready    <none>          142m   v1.25.6   10.57.172.221   <none>        Ubuntu 22.04.1 LTS   5.15.0-58-generic   containerd://1.6.15
worker3   Ready    <none>          142m   v1.25.6   10.57.172.145   <none>        Ubuntu 22.04.1 LTS   5.15.0-58-generic   containerd://1.6.15

Now we need dashboard for k8s cluster

All actions deal on local workstation
Create new service with defenition from remote repository

# create service
kubectl apply -f https://raw.githubusercontent.com/kubernetes/dashboard/v2.7.0/aio/deploy/recommended.yaml

# run proxy service
kubectl proxy

Now we can connect to dashboard from localhost station

Create admin account for access to dashboard

First time we should be create service account with name admin-user in namespace kubernetes-dashboard (default created for dashboard)
Create ./manifests/accounting/service-account.yaml

apiVersion: v1
kind: ServiceAccount
metadata:
  name: admin-user
  namespace: kubernetes-dashboard

Apply manifest

kubectl apply -f ~/cloud-init/manifests/accounting/service-account.yaml

Output:

serviceaccount/admin-user created

The most popular tools already have ClusterRole with user cluster-admin in the cluster. We need to link created user to exists admin account manualy Create ./manifests/accounting/cluster-role-binding.yaml

apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: admin-user
roleRef:
  apiGroup: rbac.authorization.k8s.io
  kind: ClusterRole
  name: cluster-admin
subjects:
- kind: ServiceAccount
  name: admin-user
  namespace: kubernetes-dashboard

Apply manifest

kubectl apply -f ~/cloud-init/manifests/accounting/cluster-role-binding.yaml

Output:

clusterrolebinding.rbac.authorization.k8s.io/admin-user created

Create admin token

kubectl -n kubernetes-dashboard create token admin-user

Output:

eyJhbGciOiJSUzI1NiIsImtpZCI6InBDNnhmUWR1VzJ6ZUVKZHdjSmgzdVVXaTA5bkU3cTUxQ1A3WkNNODY2c1EifQ.eyJhdWQiOlsiaHR0cHM6Ly9rdWJlcm5ldGVzLmRlZmF1bHQuc3ZjLmNsdXN0ZXIubG9jYWwiXSwiZXhwIjoxNjc0NTU2MjEzLCJpYXQiOjE2NzQ1NTI2MTMsImlzcyI6Imh0dHBzOi8va3ViZXJuZXRlcy5kZWZhdWx0LnN2Yy5jbHVzdGVyLmxvY2FsIiwia3ViZXJuZXRlcy5pbyI6eyJuYW1lc3BhY2UiOiJrdWJlcm5ldGVzLWRhc2hib2FyZCIsInNlcnZpY2VhY2NvdW50Ijp7Im5hbWUiOiJhZG1pbi11c2VyIiwidWlkIjoiYmE4Y2ZmMjgtOTRjNy00ZWFmLWE2NWUtNzk1ZDI0ZjUyOGZjIn19LCJuYmYiOjE2NzQ1NTI2MTMsInN1YiI6InN5c3RlbTpzZXJ2aWNlYWNjb3VudDprdWJlcm5ldGVzLWRhc2hib2FyZDphZG1pbi11c2VyIn0.qUwno1PCTepp4bzqPKyhuAO7B9cd1jg6eLNVTD8P9utr8EToHECpYSs7KnC-dJRo0L8DTydUKoHMSWZLA8SGyO7WZ4UfjeLcXDuU9vWXHOw0SnDA_FjcCP8QQUYejkTzcTGxAt73B9jOG8mm7ZIVYJkowYFotGSoQvtpb_3l0kEi9Hd0SRXh-yhjPI-BtHSBpk9rgYEwwp_sZU0wptbeli7tBNWzyrEK2ZNDKmV7VqWtvkJxaICZRkEz1PyHFcs5juslC5loRd39myaZ6fTshvivgSMPJivrWNGDWi5ycnWhpCbsXhyTDEvs7lTTx79v8eugD2m7eVX7dvLzlyLZtg

Paste output token into form and u are connected to dashboard

Choose ingress controller for access from non-demilitarized zone

So. We can choose between 3 most popular Ingress controllers and one default k8s controller:

  1. K8s default controller
  2. NGINX Ingress controller
  3. HAProxy Ingress controller
  4. Traefik Ingress controller

Default k8s and NGINX controllers are very slow that Traefik and HAProxy.
HAProxy faster that Traefik in ops on one CPU, Traefik faster that HAProxy on latency between controller and services.
I choose Traefik Ingress Controller on my learning.

For configuring Traefik and others packages we should install Helm packages manager

Fully installing introduction

# Deal all actions on local workstation
# create helm dir
mkdir ~/cloud-init/helm

# get latest release from git repo and save into helm installation dir
wget https://get.helm.sh/helm-v3.11.0-linux-amd64.tar.gz -O ~/cloud-init/helm/helm-v3.11.0-linux-amd64.tar.gz

# unpack archive
tar -zxvf ~/cloud-init/helm/helm-v3.11.0-linux-amd64.tar.gz -c ~/cloud-init/helm

# move binary 
sudo mv ~/cloud-init/helm/linux-amd64/helm /usr/local/bin/helm

# remove temporary files
rm -rf ~/cloud-init/helm

# test them
helm help

Configure Traefik Ingress Controller

Traefik schema

Instructions:

  1. Helm
  2. Native configuration