kubespray (CentOS 8)

refer to

---

• https://github.com/kubernetes-sigs/kubespray

Deploy

---

[root@kf01_deploy_0 ~]# yum -y install python3-pip wget git vim epel-release
[root@kf01_deploy_0 ~]# yum -y install sshpass

-

[root@kf01_deploy_0 ~]# git clone https://github.com/kubernetes-sigs/kubespray.git
[root@kf01_deploy_0 ~]# cd kubespray/

-

[root@kf01_deploy_0 kubespray]# pip3 install -r requirements.txt
[root@kf01_deploy_0 kubespray]# cp -rfp inventory/sample inventory/mycluster

inventory 파일 수정 ssh key로 진행해도 되나 password로 진행함

만약 key를 배포한다면 아래 명령을 사용하면 좀 편하게 할 수 있다.

[root@ha02_master_0 kubespray]# grep 'ansible_host=' inventory/mycluster/inventory.ini |grep -v '#' |awk '{print $2}' |sed 's/ansible_host=/ssh-copy-id /'
ssh-copy-id 10.96.90.10
ssh-copy-id 10.96.90.11
ssh-copy-id 10.96.90.12
ssh-copy-id 10.96.90.20
ssh-copy-id 10.96.90.21

inventory 제일 앞에 있는 node의 이름이 각 node의 hostname이됨

# 환경에 맞게 수정
[root@kf01_deploy_0 kubespray]# vim inventory/mycluster/inventory.ini

# ## Configure 'ip' variable to bind kubernetes services on a
# ## different ip than the default iface
# ## We should set etcd_member_name for etcd cluster. The node that is not a etcd member do not need to set the value, or can set the empty string value.
[all]
master01 ansible_host=10.13.90.10 ip=10.13.62.10 etcd_member_name=etcd1
master02 ansible_host=10.13.90.11 ip=10.13.62.11 
worker01 ansible_host=10.13.90.20 ip=10.13.62.20 
worker02 ansible_host=10.13.90.21 ip=10.13.62.21 

# ## configure a bastion host if your nodes are not directly reachable
# bastion ansible_host=x.x.x.x ansible_user=some_user

[kube-master]
master01
master02

[etcd]
master01

[kube-node]
worker01
worker02

[calico-rr]

[k8s-cluster:children]
kube-master
kube-node
calico-rr

[all:vars]
ansible_connection=ssh
ansible_user=root
ansible_ssh_pass=testtest

[root@kf01_deploy_0 kubespray]# ansible -m ping -i inventory/mycluster/inventory.ini all
worker02 | SUCCESS => {
    "ansible_facts": {
        "discovered_interpreter_python": "/usr/libexec/platform-python"
    },
    "changed": false,
    "ping": "pong"
}
master01 | SUCCESS => {
    "ansible_facts": {
        "discovered_interpreter_python": "/usr/libexec/platform-python"
    },
    "changed": false,
    "ping": "pong"
}
master02 | SUCCESS => {
    "ansible_facts": {
        "discovered_interpreter_python": "/usr/libexec/platform-python"
    },
    "changed": false,
    "ping": "pong"
}
worker01 | SUCCESS => {
    "ansible_facts": {
        "discovered_interpreter_python": "/usr/libexec/platform-python"
    },
    "changed": false,
    "ping": "pong"
}

-

# 환경에 맞게 수정
[root@kf01_deploy_0 kubespray]# vim inventory/mycluster/group_vars/all/all.yml
[root@kf01_deploy_0 kubespray]# vim inventory/mycluster/group_vars/k8s-cluster/k8s-cluster.yml

Cluster 배포 진행

[root@kf01_deploy_0 kubespray]# ansible-playbook -i inventory/mycluster/inventory.ini cluster.yml 
...
PLAY RECAP *********************************************************************************************************************************
localhost                  : ok=1    changed=0    unreachable=0    failed=0    skipped=0    rescued=0    ignored=0   
master01                   : ok=557  changed=121  unreachable=0    failed=0    skipped=1066 rescued=0    ignored=0   
master02                   : ok=413  changed=93   unreachable=0    failed=0    skipped=874  rescued=0    ignored=0   
worker01                   : ok=349  changed=75   unreachable=0    failed=0    skipped=569  rescued=0    ignored=0   
worker02                   : ok=349  changed=75   unreachable=0    failed=0    skipped=568  rescued=0    ignored=0   
...

Cluster 배포 확인 (master 서버에서 진행할 예정)

[root@master01 ~]# kubectl get all -A
NAMESPACE     NAME                                          READY   STATUS    RESTARTS   AGE
kube-system   pod/calico-kube-controllers-8b5ff5d58-dnxff   1/1     Running   0          2m10s
kube-system   pod/calico-node-7ngwv                         1/1     Running   0          2m55s
kube-system   pod/calico-node-8jd9z                         1/1     Running   0          2m55s
kube-system   pod/calico-node-9zhwr                         1/1     Running   0          2m55s
...

[root@master01 ~]# kubectl cluster-info
Kubernetes master is running at https://10.13.62.10:6443

[root@master01 ~]# kubectl config view
apiVersion: v1
clusters:
- cluster:
    certificate-authority-data: DATA+OMITTED
    server: https://10.13.62.10:6443
  name: k8s.yjwang
contexts:
- context:
    cluster: k8s.yjwang
    user: kubernetes-admin
  name: kubernetes-admin@k8s.yjwang
current-context: kubernetes-admin@k8s.yjwang
kind: Config
preferences: {}
users:
- name: kubernetes-admin
  user:
    client-certificate-data: REDACTED
    client-key-data: REDACTED

bash completion 적용

[root@master01 ~]# yum -y install bash-completion
[root@master01 ~]# kubectl completion bash > /etc/bash_completion.d/kubectl
[root@master01 ~]# exit

# 재 로그인

[root@master01 ~]# kubectl get [tab]
apiservices.apiregistration.k8s.io                            mutatingwebhookconfigurations.admissionregistration.k8s.io
certificatesigningrequests.certificates.k8s.io                namespaces
clusterrolebindings.rbac.authorization.k8s.io                 networkpolicies.networking.k8s.io
...

[option] addon 및 metallb 추가하기

---

[root@addon01_deploy_0 ~]# grep -v ^# kubespray/inventory/addon/group_vars/k8s-cluster/addons.yml 
---
dashboard_enabled: true

helm_enabled: true

...

metallb_enabled: true
metallb_ip_range:
   - "10.18.62.50-10.18.62.99"
metallb_version: v0.9.3
metallb_protocol: "layer2"
metallb_port: "7472"

-

[root@addon01_deploy_0 ~]# grep strict kubespray/inventory/addon/group_vars/k8s-cluster/k8s-cluster.yml 
kube_proxy_strict_arp: true