1、安裝Docker
在21、22、200三臺機器上安裝Docker。安裝命令:
在21、22、200三臺主機上部署Docker。
~]# curl -fsSL https://get.docker.com | bash -s docker --mirror Aliyun
1.1 配置Docker
/etc/docker/daemon.json
{
"graph": "/data/docker",
"storage-driver": "overlay2",
"insecure-registries": ["registry.access.redhat.com","quay.io","harbor.od.com"],
"registry-mirrors": ["https://q2gr04ke.mirror.aliyuncs.com"],
"bip": "172.7.21.1/24",
"exec-opts": ["native.cgroupdriver=systemd"],
"live-restore": true
}
說明:
exec-opts:CPU/MEM的資源管理方式
registry-mirrors:映象源
insecure-registries:信任的HTTP映象倉庫
bip根據不同的主機修改:
21:172.7.21.1/24
22:172.7.22.1/24
200: 172.7.200.1/24
建立目錄
[root@hdss7-21 ~]# mkdir -pv /data/docker
mkdir: created directory ‘/data’
mkdir: created directory ‘/data/docker’
[root@hdss7-21 ~]#
1.2 啟動docker
~]# systemctl enable docker
~]# systemctl start docker
~]# systemctl status docker -l
~]# docker info
~]# docker version
2 部署kube-apiserver叢集
2.1 叢集規劃
主機名 | 角色 | IP |
---|---|---|
CFZX55-21.host.com | kube-apiserver | 10.211.55.21 |
CFZX55-22.host.com | kube-apiserver | 10.211.55.22 |
CFZX55-11.host.com | 4層負載均衡 Nginx+Keepalived | 10.211.55.11 |
CFZX55-12.host.com | 4層負載均衡 Nginx+Keepalived | 10.211.55.12 |
注意:這裡10.211.55.11和10.211.55.12使用nginx做4層負載均衡,用keepalived跑一個vip:10.211.55.10,代理兩個kube-apiserver,實現高可用。
2.2 下載軟體、解壓、做軟連結
在21主機上操作
本例使用1.23.4版本。
[root@cfzx55-21 src]# tar xf kubernetes-server-linux-amd64.tar.gz -C /opt/
[root@cfzx55-21 src]# cd ..
[root@cfzx55-21 kubernetes]# rm kubernetes-src.tar.gz -f
[root@cfzx55-21 kubernetes]# rm -rf LICENSES/
[root@cfzx55-21 kubernetes]# rm -rf addons/
[root@cfzx55-21 kubernetes]# cd server/
[root@cfzx55-21 server]# mv bin/ ../
[root@cfzx55-21 server]# cd ..
[root@cfzx55-21 kubernetes]# rm -rf server/
[root@cfzx55-21 kubernetes]# cd bin/
[root@cfzx55-21 bin]# rm *.tar -f
[root@cfzx55-21 bin]# rm *_tag -f
[root@cfzx55-21 opt]# vim /etc/profile
export PATH=$PATH:/opt/etcd:/opt/kubernetes/bin
[root@cfzx55-21 opt]# source /etc/profile
2.3 簽發kube-apiserver證書
在200主機上操作
/opt/certs/kube-apiserver-csr.json
{
"CN": "kube-apiserver",
"hosts": [
"127.0.0.1",
"192.168.0.1",
"kubernetes",
"kubernetes.default",
"kubernetes.default.svc",
"kubernetes.default.svc.cluster",
"kubernetes.default.svc.cluster.local",
"10.211.55.10",
"10.211.55.21",
"10.211.55.22",
"10.211.55.23"
],
"key": {
"algo": "rsa",
"size": 2048
},
"names": [
{
"C": "CN",
"ST": "beijing",
"L": "beijing",
"O": "system:masters",
"OU": "system"
}
]
}
說明:
- CN:K8S會提取CN欄位的值作為使用者名稱,實際是指K8S的"RoleBinding/ClusterRoleBinding"資源中,“subjects.kind”的值為“User",
- hosts:包括所有Master節點的IP地址,LB節點、LB叢集節點、ClusterIP的首個IP,K8S的“ClusterIP”的範圍在“--service-cluster-ip-range”中指定,取值為192.168.0.0/16,此處配置為192.168.0.1
- names
- C:CN
- ST:
- L:
- O:“system:masters”,定義“O”值的原因:apiserver向kubelet發起請求時,將複用此證書,參看官方文件。K8S預設會提取“O”欄位的值作為組,這實際是指K8S的“RoleBinding/ClusterRoleBinding”資源中"subjects.kind"的值為“Group”,
生成證書
[root@cfzx55-200 certs]# cfssl gencert \
> -ca=ca.pem \
> -ca-key=ca-key.pem \
> -config=ca-config.json \
> -profile=kubernetes \
> kube-apiserver-csr.json | cfssl-json -bare kube-apiserver
2022/03/12 21:28:43 [INFO] generate received request
2022/03/12 21:28:43 [INFO] received CSR
2022/03/12 21:28:43 [INFO] generating key: rsa-2048
2022/03/12 21:28:43 [INFO] encoded CSR
2022/03/12 21:28:43 [INFO] signed certificate with serial number 218531774642654852589087643914770351081106577228
[root@cfzx55-200 certs]# ll kube-apiserver*
-rw-r--r-- 1 root root 636 Mar 12 21:27 kube-apiserver-csr.json
-rw------- 1 root root 1679 Mar 12 21:28 kube-apiserver-key.pem
-rw-r--r-- 1 root root 1289 Mar 12 21:28 kube-apiserver.csr
-rw-r--r-- 1 root root 1655 Mar 12 21:28 kube-apiserver.pem
[root@cfzx55-200 certs]#
2.4 拷貝證書至各運算節點
在21主機上操作,把6張證書和金鑰從200主機上拷貝到certs目錄下
[root@cfzx55-21 certs]# pwd
/opt/kubernetes/certs
[root@cfzx55-21 certs]# ll
total 24
-rw------- 1 root root 1679 Mar 12 21:32 ca-key.pem
-rw-r--r-- 1 root root 1310 Mar 12 21:32 ca.pem
-rw------- 1 root root 1675 Mar 12 21:32 etcd-key.pem
-rw-r--r-- 1 root root 1448 Mar 12 21:32 etcd.pem
-rw------- 1 root root 1679 Mar 12 21:32 kube-apiserver-key.pem
-rw-r--r-- 1 root root 1655 Mar 12 21:32 kube-apiserver.pem
[root@cfzx55-21 certs]#
2.5 生成token.csv檔案
該檔案的作用是,在工作節點(kubelet)加入K8S叢集的過程中,向kube-apiserver申請簽發證書。
/opt/kubernetes/bin/certs/kube-apiserver.token.csv
[root@cfzx55-21 certs]# head -c 16 /dev/urandom | od -An -t x | tr -d " "
cceb7b589306a60ab6afe922f1f32d50
[root@cfzx55-21 certs]# echo cceb7b589306a60ab6afe922f1f32d50,kubelet-bootstrap,10001,"system:kubelet-bootstrap" > kube-apiserver.token.csv
[root@cfzx55-21 certs]# cat kube-apiserver.token.csv
cceb7b589306a60ab6afe922f1f32d50,kubelet-bootstrap,10001,system:kubelet-bootstrap
[root@cfzx55-21 certs]#
2.6 建立啟動指令碼
在21主機上操作
建立啟動指令碼
/opt/kubernetes/bin/kube-apiserver-startup.sh
#!/bin/bash
./kube-apiserver \
--runtime-config=api/all=true \
--anonymous-auth=false \
--bind-address=0.0.0.0 \
--advertise-address=10.211.55.21 \
--secure-port=6443 \
--tls-cert-file=./certs/kube-apiserver.pem \
--tls-private-key-file=./certs/kube-apiserver-key.pem \
--client-ca-file=./certs/ca.pem \
--etcd-cafile=./certs/ca.pem \
--etcd-certfile=./certs/etcd.pem \
--etcd-keyfile=./certs/etcd-key.pem \
--etcd-servers=https://10.211.55.12:2379,https://10.211.55.21:2379,https://10.211.55.22:2379 \
--kubelet-client-certificate=./certs/kube-apiserver.pem \
--kubelet-client-key=./certs/kube-apiserver-key.pem \
--service-account-key-file=./certs/ca.pem \
--service-account-signing-key-file=./certs/ca-key.pem \
--service-account-issuer=https://kubernetes.default.svc.cluster.local \
--enable-bootstrap-token-auth=true \
--token-auth-file=./certs/kube-apiserver.token.csv \
--allow-privileged=true \
--service-cluster-ip-range=192.168.0.0/16 \
--service-node-port-range=8000-20000 \
--authorization-mode=RBAC,Node \
--enable-aggregator-routing=true \
--enable-admission-plugins=NamespaceLifecycle,LimitRanger,ServiceAccount,ResourceQuota,NodeRestriction \
--v=2 \
--audit-log-path=/data/logs/kubernetes/kube-apiserver/audit-log \
--log-dir=/data/logs/kubernetes/kube-apiserver
2.7 調整許可權和目錄
[root@cfzx55-21 bin]# chmod +x kube-apiserver-startup.sh
[root@cfzx55-21 bin]# mkdir -pv /data/logs/kubernetes/kube-apiserver
2.8 建立supervisor配置
/etc/supervisord.d/kube-apiserver.ini
[program:kube-apiserver-55-21]
command=/opt/kubernetes/bin/kube-apiserver-startup.sh
numprocs=1
directory=/opt/kubernetes/bin
autostart=true
autorestart=true
startsecs=30
startretries=3
exitcodes=0,2
stopsignal=QUIT
stopwaitsecs=10
user=root
redirect_stderr=true
stdout_logfile=/data/logs/kubernetes/kube-apiserver/apiserver.stdout.log
stdout_logfile_maxbytes=64MB
stdout_logfile_backups=5
stdout_capture_maxbytes=1MB
stdout_events_enabled=false
2.9 啟動服務並檢查
[root@cfzx55-21 bin]# supervisorctl update
[root@cfzx55-21 bin]# supervisorctl status
etcd-server-55-21 RUNNING pid 12536, uptime 2:29:07
kube-apiserver-55-21 RUNNING pid 13122, uptime 0:00:40
[root@cfzx55-21 bin]# netstat -luntp
Active Internet connections (only servers)
Proto Recv-Q Send-Q Local Address Foreign Address State PID/Program name
tcp 0 0 0.0.0.0:6443 0.0.0.0:* LISTEN 13123/./kube-apiser
tcp 0 0 10.211.55.21:2379 0.0.0.0:* LISTEN 12537/./etcd
tcp 0 0 127.0.0.1:2379 0.0.0.0:* LISTEN 12537/./etcd
tcp 0 0 10.211.55.21:2380 0.0.0.0:* LISTEN 12537/./etcd
tcp 0 0 10.211.55.21:2381 0.0.0.0:* LISTEN 12537/./etcd
tcp 0 0 0.0.0.0:22 0.0.0.0:* LISTEN 912/sshd
udp 0 0 127.0.0.1:323 0.0.0.0:* 700/chronyd
[root@cfzx55-21 bin]#
2.10 安裝部署所有節點
安裝22節點
# 在21節點上,把kubernetes檔案拷貝到22節點
[root@cfzx55-21 ~]# scp -r /opt/kubernetes/ root@cfzx55-22:/opt/
[root@cfzx55-21 ~]# scp /etc/supervisord.d/kube-apiserver.ini root@cfzx55-22:/etc/supervisord.d/
# 在22節點上,建立目錄
[root@cfzx55-22 certs]# mkdir -pv /data/logs/kubernetes/kube-apiserver
# 修改 kube-apiserver-startup.sh 中的ip地址
# 修改 /etc/supervisord.d/kube-apiserver.ini 中的名稱
# 啟動
[root@cfzx55-22 bin]# supervisorctl update
[root@cfzx55-22 bin]# supervisorctl status
etcd-server-55-22 RUNNING pid 12495, uptime 2:37:27
kube-apiserver-55-22 RUNNING pid 12675, uptime 0:00:38
[root@cfzx55-22 bin]# netstat -luntp
Active Internet connections (only servers)
Proto Recv-Q Send-Q Local Address Foreign Address State PID/Program name
tcp 0 0 0.0.0.0:6443 0.0.0.0:* LISTEN 12676/./kube-apiser
tcp 0 0 10.211.55.22:2379 0.0.0.0:* LISTEN 12496/./etcd
tcp 0 0 127.0.0.1:2379 0.0.0.0:* LISTEN 12496/./etcd
tcp 0 0 10.211.55.22:2380 0.0.0.0:* LISTEN 12496/./etcd
tcp 0 0 10.211.55.22:2381 0.0.0.0:* LISTEN 12496/./etcd
tcp 0 0 0.0.0.0:22 0.0.0.0:* LISTEN 914/sshd
udp 0 0 127.0.0.1:323 0.0.0.0:* 704/chronyd
[root@cfzx55-22 bin]#
檢查叢集狀態
[root@cfzx55-21 ~]# curl --insecure https://10.211.55.21:6443/
{
"kind": "Status",
"apiVersion": "v1",
"metadata": {},
"status": "Failure",
"message": "Unauthorized",
"reason": "Unauthorized",
"code": 401
}[root@cfzx55-21 ~]# curl --insecure https://10.211.55.22:6443/
{
"kind": "Status",
"apiVersion": "v1",
"metadata": {},
"status": "Failure",
"message": "Unauthorized",
"reason": "Unauthorized",
"code": 401
}[root@cfzx55-21 ~]#
至此,kube-apiserver安裝完成。
2.11 配置4層反向代理
apiserver監聽埠
[root@cfzx55-21 ~]# netstat -luntp | grep kube
tcp 0 0 0.0.0.0:6443 0.0.0.0:* LISTEN 13123/./kube-apiser
[root@cfzx55-22 ~]# netstat -luntp | grep kube
tcp 0 0 0.0.0.0:6443 0.0.0.0:* LISTEN 12676/./kube-apiser
用keepalived跑一個10.211.55.10的vip
用10.211.55.10上的7443,反向代理10.211.55.21和10.211.55.22上的6443埠。
下面的操作在11和12主機上進行。
安裝nginx
~]# yum install nginx -y
~]# yum install nginx-mod-stream -y
配置nginx
把下面內容,新增到/etc/nginx/nginx.conf
檔案最後,也就是並列在http模組的後面。
stream {
upstream kube-apiserver {
server 10.211.55.21:6443 max_fails=3 fail_timeout=30s;
server 10.211.55.22:6443 max_fails=3 fail_timeout=30s;
}
server {
listen 7443;
proxy_connect_timeout 2s;
proxy_timeout 900s;
proxy_pass kube-apiserver;
}
}
[root@cfzx55-11 ~]# nginx -t
nginx: [emerg] unknown directive "stream" in /etc/nginx/nginx.conf:85
nginx: configuration file /etc/nginx/nginx.conf test failed
上面錯誤提示,是因為沒有安裝stream模組。
啟動nginx
~]# systemctl start nginx
~]# systemctl enable nginx
~]# systemctl status nginx
檢查狀態
[root@cfzx55-11 ~]# netstat -luntp | grep nginx
tcp 0 0 0.0.0.0:80 0.0.0.0:* LISTEN 1499/nginx: master
tcp 0 0 0.0.0.0:7443 0.0.0.0:* LISTEN 1499/nginx: master
[root@cfzx55-11 ~]#
安裝keepalived
[root@cfzx55-11 ~]# yum install keepalived -y
[root@hdss7-12 ~]# yum install keepalived -y
建立監聽指令碼
/etc/keepalived/check_port.sh
#!/bin/bash
CHK_PORT=$1
if [ -n "$CHK_PORT" ];then
PORT_PROCESS=`ss -lnt | grep $CHK_PORT | wc -l`
if [ $PORT_PROCESS -eq 0 ];then
echo "Port $CHK_PORT Is Not Used, End."
exit 1
fi
else
echo "Check Port Cant Be Empty!"
fi
增加執行許可權
~]# chmod +x /etc/keepalived/check_port.sh
測試指令碼
[root@cfzx55-11 ~]# /etc/keepalived/check_port.sh
Check Port Cant Be Empty!
[root@cfzx55-11 ~]# /etc/keepalived/check_port.sh 7443
[root@cfzx55-11 ~]# echo $?
0
[root@cfzx55-11 ~]# /etc/keepalived/check_port.sh 7445
Port 7445 Is Not Used, End.
[root@cfzx55-11 ~]# echo $?
1
[root@cfzx55-11 ~]#
keepalived主配置
在11主機上操作
/etc/keepalived/keepalived.conf
! Configuration File for keepalived
global_defs {
router_id 10.211.55.11
}
vrrp_script chk_nginx {
script "/etc/keepalived/check_port.sh 7443"
interval 2
weight -20
}
vrrp_instance VI_1 {
state MASTER
interface eth0
virtual_router_id 251
priority 100
advert_int 1
mcast_src_ip 10.211.55.11
nopreempt #設定非搶佔式,當主服務down,vip漂移到備機,當主機服務up,vip依然在備機上
authentication {
auth_type PASS
auth_pass 11111111
}
track_script {
chk_nginx
}
virtual_ipaddress {
10.211.55.10
}
}
keepalived從配置
在12主機上操作
/etc/keepalived/keepalived.conf
! Configuration File for keepalived
global_defs {
router_id 10.211.55.12
}
vrrp_script chk_nginx {
script "/etc/keepalived/check_port.sh 7443"
interval 2
weight -20
}
vrrp_instance VI_1 {
state BACKUP
interface eth0
virtual_router_id 251
priority 90
advert_int 1
mcast_src_ip 10.211.55.12
! 注意 備機上不能有 nopreempt 配置
authentication {
auth_type PASS
auth_pass 11111111
}
track_script {
chk_nginx
}
virtual_ipaddress {
10.211.55.10
}
}
2.12 啟動代理並檢查
在11主機上操作
[root@cfzx55-11 ~]# systemctl enable keepalived
Created symlink from /etc/systemd/system/multi-user.target.wants/keepalived.service to /usr/lib/systemd/system/keepalived.service.
[root@cfzx55-11 ~]# systemctl start keepalived
[root@cfzx55-11 ~]# systemctl status keepalived
● keepalived.service - LVS and VRRP High Availability Monitor
Loaded: loaded (/usr/lib/systemd/system/keepalived.service; enabled; vendor preset: disabled)
Active: active (running) since Sun 2022-03-13 08:21:53 CST; 6s ago
Process: 1580 ExecStart=/usr/sbin/keepalived $KEEPALIVED_OPTIONS (code=exited, status=0/SUCCESS)
Main PID: 1581 (keepalived)
CGroup: /system.slice/keepalived.service
├─1581 /usr/sbin/keepalived -D
├─1582 /usr/sbin/keepalived -D
└─1583 /usr/sbin/keepalived -D
Mar 13 08:21:53 cfzx55-11.host.com Keepalived_healthcheckers[1582]: Opening file '/etc/keepalive....
Mar 13 08:21:54 cfzx55-11.host.com Keepalived_vrrp[1583]: VRRP_Instance(VI_1) Transition to MAS...TE
Mar 13 08:21:55 cfzx55-11.host.com Keepalived_vrrp[1583]: VRRP_Instance(VI_1) Entering MASTER STATE
Mar 13 08:21:55 cfzx55-11.host.com Keepalived_vrrp[1583]: VRRP_Instance(VI_1) setting protocol VIPs.
Mar 13 08:21:55 cfzx55-11.host.com Keepalived_vrrp[1583]: Sending gratuitous ARP on eth0 for 10...10
Mar 13 08:21:55 cfzx55-11.host.com Keepalived_vrrp[1583]: VRRP_Instance(VI_1) Sending/queueing ...10
Mar 13 08:21:55 cfzx55-11.host.com Keepalived_vrrp[1583]: Sending gratuitous ARP on eth0 for 10...10
Mar 13 08:21:55 cfzx55-11.host.com Keepalived_vrrp[1583]: Sending gratuitous ARP on eth0 for 10...10
Mar 13 08:21:55 cfzx55-11.host.com Keepalived_vrrp[1583]: Sending gratuitous ARP on eth0 for 10...10
Mar 13 08:21:55 cfzx55-11.host.com Keepalived_vrrp[1583]: Sending gratuitous ARP on eth0 for 10...10
Hint: Some lines were ellipsized, use -l to show in full.
[root@cfzx55-11 ~]#
在12主機上操作
[root@cfzx55-12 ~]# systemctl enable keepalived
Created symlink from /etc/systemd/system/multi-user.target.wants/keepalived.service to /usr/lib/systemd/system/keepalived.service.
[root@cfzx55-12 ~]# systemctl start keepalived
[root@cfzx55-12 ~]# systemctl status keepalived
● keepalived.service - LVS and VRRP High Availability Monitor
Loaded: loaded (/usr/lib/systemd/system/keepalived.service; enabled; vendor preset: disabled)
Active: active (running) since Sun 2022-03-13 08:22:29 CST; 7s ago
Process: 1538 ExecStart=/usr/sbin/keepalived $KEEPALIVED_OPTIONS (code=exited, status=0/SUCCESS)
Main PID: 1539 (keepalived)
CGroup: /system.slice/keepalived.service
├─1539 /usr/sbin/keepalived -D
├─1540 /usr/sbin/keepalived -D
└─1541 /usr/sbin/keepalived -D
Mar 13 08:22:29 cfzx55-12.host.com Keepalived_vrrp[1541]: Registering gratuitous ARP shared channel
Mar 13 08:22:29 cfzx55-12.host.com Keepalived_vrrp[1541]: Opening file '/etc/keepalived/keepali...'.
Mar 13 08:22:29 cfzx55-12.host.com Keepalived_vrrp[1541]: WARNING - default user 'keepalived_sc...e.
Mar 13 08:22:29 cfzx55-12.host.com Keepalived_vrrp[1541]: Unable to access script `/etc/keepali...h`
Mar 13 08:22:29 cfzx55-12.host.com Keepalived_vrrp[1541]: Disabling track script chk_nginx sinc...nd
Mar 13 08:22:29 cfzx55-12.host.com Keepalived_vrrp[1541]: VRRP_Instance(VI_1) removing protocol...s.
Mar 13 08:22:29 cfzx55-12.host.com Keepalived_vrrp[1541]: Using LinkWatch kernel netlink reflec.....
Mar 13 08:22:29 cfzx55-12.host.com Keepalived_vrrp[1541]: VRRP_Instance(VI_1) Entering BACKUP STATE
Mar 13 08:22:29 cfzx55-12.host.com Keepalived_vrrp[1541]: VRRP sockpool: [ifindex(2), proto(112...)]
Mar 13 08:22:29 cfzx55-12.host.com Keepalived_healthcheckers[1540]: Opening file '/etc/keepalive....
Hint: Some lines were ellipsized, use -l to show in full.
[root@cfzx55-12 ~]#
在11上用ip addr
命令,能看到VIP
[root@cfzx55-11 ~]# ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc
link/ether 00:1c:42:76:65:e1 brd ff:ff:ff:ff:ff:ff
inet 10.211.55.11/24 brd 10.211.55.255 scope global noprefixroute eth0
valid_lft forever preferred_lft forever
inet 10.211.55.10/32 scope global eth0
valid_lft forever preferred_lft forever
[root@cfzx55-11 ~]#
Nginx+Keepalived高可用測試
在11主機上,停止nginx,vip不存在了。
[root@cfzx55-11 ~]# systemctl stop nginx
[root@cfzx55-11 ~]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc
link/ether 00:1c:42:76:65:e1 brd ff:ff:ff:ff:ff:ff
inet 10.211.55.11/24 brd 10.211.55.255 scope global noprefixroute eth0
valid_lft forever preferred_lft forever
[root@cfzx55-11 ~]#
12主機上檢視,vip跑在了12主機上
[root@cfzx55-12 ~]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc
link/ether 00:1c:42:e2:45:7b brd ff:ff:ff:ff:ff:ff
inet 10.211.55.12/24 brd 10.211.55.255 scope global noprefixroute eth0
valid_lft forever preferred_lft forever
inet 10.211.55.10/32 scope global eth0
valid_lft forever preferred_lft forever
[root@cfzx55-12 ~]#
重啟nginx,vip還在12主機上。
這是因為在11上配置了 nopreempt ,設定非搶佔式,當主服務down,vip漂移到備機,當主機服務up,vip依然在備機上。
如果要想vip回到11上,重新啟動keepalived。
3 部署kubectl元件
3.1 叢集規劃
主機名 | 角色 | IP |
---|---|---|
CFZX55-21.host.com | kubectl | 10.211.55.21 |
CFZX55-22.host.com | kubectl | 10.211.55.22 |
3.2 簽發kubectl證書
在運維主機200上操作
生成kubectl證書請求csr檔案
/opt/certs/kubectl-csr.json
{
"CN": "clusteradmin",
"hosts": [],
"key": {
"algo": "rsa",
"size": 2048
},
"names": [
{
"C": "CN",
"ST": "beijing",
"L": "beijing",
"O": "system:masters",
"OU": "system"
}
]
}
說明
- CN:kubectl證書中的CN值沒有意義,隨便取值
- O:因為希望使用kubectl時有完整的叢集操作許可權,所以取值為“system:masters”,K8S預設會提取O欄位的值作為組,這實際是指K8S裡“RoleBinding/ClusterRoleBinding”資源中"subjects.kind"的值為“Group”
- 後續 kube-apiserver 使用 RBAC 對客戶端(如 kubelet、kube-proxy、Pod)請求進行授權;
kube-apiserver 預定義了一些 RBAC 使用的 RoleBindings,如 cluster-admin 將 Group system:masters 與 Role cluster-admin 繫結,該 Role 授予了呼叫kube-apiserver 的所有 API的許可權;- O指定該證書的 Group 為 system:masters,kubelet 使用該證書訪問 kube-apiserver 時 ,由於證書被 CA 簽名,所以認證通過,同時由於證書使用者組為經過預授權的 system:masters,所以被授予訪問所有 API 的許可權;
- 這個證書,是將來生成管理員用的kube config 配置檔案用的,現在我們一般建議使用RBAC 來對kubernetes 進行角色許可權控制, kubernetes 將證書中的CN 欄位 作為User, O 欄位作為 Group;
"O": "system:masters", 必須是system:masters,否則後面kubectl create clusterrolebinding報錯。
生成證書
[root@cfzx55-200 certs]# cfssl gencert \
> -ca=ca.pem \
> -ca-key=ca-key.pem \
> -config=ca-config.json \
> -profile=kubernetes \
> kubectl-csr.json | cfssl-json -bare kubectl
2022/03/13 08:48:36 [INFO] generate received request
2022/03/13 08:48:36 [INFO] received CSR
2022/03/13 08:48:36 [INFO] generating key: rsa-2048
2022/03/13 08:48:36 [INFO] encoded CSR
2022/03/13 08:48:36 [INFO] signed certificate with serial number 629903670193912591906490478447930251557864868755
2022/03/13 08:48:36 [WARNING] This certificate lacks a "hosts" field. This makes it unsuitable for
websites. For more information see the Baseline Requirements for the Issuance and Management
of Publicly-Trusted Certificates, v.1.1.6, from the CA/Browser Forum (https://cabforum.org);
specifically, section 10.2.3 ("Information Requirements").
[root@cfzx55-200 certs]# ll kubectl*
-rw-r--r-- 1 root root 1017 Mar 13 08:48 kubectl.csr
-rw-r--r-- 1 root root 306 Mar 13 08:44 kubectl-csr.json
-rw------- 1 root root 1679 Mar 13 08:48 kubectl-key.pem
-rw-r--r-- 1 root root 1411 Mar 13 08:48 kubectl.pem
[root@cfzx55-200 certs]#
3.3 把證書拷貝到21和22主機上
[root@cfzx55-200 certs]# scp kubectl*.pem root@cfzx55-21:/opt/kubernetes/bin/certs/
[root@cfzx55-200 certs]# scp kubectl*.pem root@cfzx55-22:/opt/kubernetes/bin/certs/
3.4 生成kubeconfig配置檔案
生成kubectl元件的kubectl.kubeconfig配置檔案,該檔案包含訪問kube-apiseerver的所有資訊,如kube-apiserver的地址,CA證書和自身使用的證書。
有了這個檔案,便可以在任何機器上以超級管理員身份對K8S叢集做任何操作,請務必保證此檔案的安全性。
kubectl命令預設使用的配置檔案為:~/.kube.config
以下在21主機上操作,完成後把生成的檔案拷貝到其餘Master節點,本例是22主機。
生成建立配置檔案的指令碼
#!/bin/bash
KUBE_CONFIG="/root/.kube/config"
KUBE_APISERVER="https://10.211.55.10:7443"
kubectl config set-cluster kubernetes \
--certificate-authority=/opt/kubernetes/bin/certs/ca.pem \
--embed-certs=true \
--server=${KUBE_APISERVER} \
--kubeconfig=${KUBE_CONFIG}
kubectl config set-credentials clusteradmin \
--client-certificate=/opt/kubernetes/bin/certs/kubectl.pem \
--client-key=/opt/kubernetes/bin/certs/kubectl-key.pem \
--embed-certs=true \
--kubeconfig=${KUBE_CONFIG}
kubectl config set-context default \
--cluster=kubernetes \
--user=clusteradmin \
--kubeconfig=${KUBE_CONFIG}
kubectl config use-context default --kubeconfig=${KUBE_CONFIG}
說明:
- 叢集名稱:描述叢集資訊的標記,沒有實際意義。
- certificate-authority:K8S叢集的根CA證書
- server:指向kube-apiserrver負載均衡器VIP的地址
- kubeconfig:生成的kubeconfig檔案
- 使用者名稱稱:clusteradmin為定義一個使用者,在kubeconfig配置檔案中,這個使用者用於關聯一組證書,這個證書對K8S叢集來說無實際意義,真正重要的是證書中的O欄位與CN欄位的定義。
- client-certificate:客戶端證書
- client-key:客戶端私鑰
- 上下文:default用於將kubeconfig配置檔案“clusteradmin”和“kubernetes”作關聯。
- cluster:set-cluster命令配置的叢集名稱。
- cluster:set-credentials命令配置的使用者名稱稱。
執行指令碼
[root@cfzx55-21 ~]# mkdir ~/.kube
[root@cfzx55-21 ~]# mkdir k8s-shell
[root@cfzx55-21 ~]# cd k8s-shell/
[root@cfzx55-21 k8s-shell]# vim kubectl-config.sh
[root@cfzx55-21 k8s-shell]# chmod +x kubectl-config.sh
[root@cfzx55-21 k8s-shell]# ./kubectl-config.sh
Cluster "kubernetes" set.
User "clusteradmin" set.
Context "default" created.
Switched to context "default".
[root@cfzx55-21 k8s-shell]#
檢視叢集狀態
[root@cfzx55-21 k8s-shell]# kubectl cluster-info
Kubernetes control plane is running at https://10.211.55.10:7443
To further debug and diagnose cluster problems, use 'kubectl cluster-info dump'.
[root@cfzx55-21 k8s-shell]# kubectl get componentstatuses
Warning: v1 ComponentStatus is deprecated in v1.19+
NAME STATUS MESSAGE ERROR
controller-manager Unhealthy Get "https://127.0.0.1:10257/healthz": dial tcp 127.0.0.1:10257: connect: connection refused
scheduler Unhealthy Get "https://127.0.0.1:10259/healthz": dial tcp 127.0.0.1:10259: connect: connection refused
etcd-1 Healthy {"health":"true","reason":""}
etcd-2 Healthy {"health":"true","reason":""}
etcd-0 Healthy {"health":"true","reason":""}
[root@cfzx55-21 k8s-shell]# kubectl get all -A
NAMESPACE NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
default service/kubernetes ClusterIP 192.168.0.1 <none> 443/TCP 10h
[root@cfzx55-21 k8s-shell]#
把21主機上生成的kubeconfig檔案拷貝到22節點
[root@cfzx55-22 ~]# mkdir ~/.kube
[root@cfzx55-22 ~]# scp root@cfzx55-21:/root/.kube/config ~/.kube/
[root@cfzx55-22 ~]# ll .kube/
total 8
-rw------- 1 root root 6224 Mar 13 09:42 config
[root@cfzx55-22 ~]#
在22上檢視叢集狀態
[root@cfzx55-22 ~]# kubectl cluster-info
Kubernetes control plane is running at https://10.211.55.10:7443
To further debug and diagnose cluster problems, use 'kubectl cluster-info dump'.
[root@cfzx55-22 ~]# kubectl get componentstatuses
Warning: v1 ComponentStatus is deprecated in v1.19+
NAME STATUS MESSAGE ERROR
controller-manager Unhealthy Get "https://127.0.0.1:10257/healthz": dial tcp 127.0.0.1:10257: connect: connection refused
scheduler Unhealthy Get "https://127.0.0.1:10259/healthz": dial tcp 127.0.0.1:10259: connect: connection refused
etcd-2 Healthy {"health":"true","reason":""}
etcd-1 Healthy {"health":"true","reason":""}
etcd-0 Healthy {"health":"true","reason":""}
[root@cfzx55-22 ~]# kubectl get all -A
NAMESPACE NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
default service/kubernetes ClusterIP 192.168.0.1 <none> 443/TCP 11h
[root@cfzx55-22 ~]#
至此,kubectl主機部署完成。
4 部署controller-manager
4.1叢集規劃
主機名 | 角色 | IP |
---|---|---|
CFZX55-21.host.com | controller-manager | 10.211.55.21 |
CFZX55-22.host.com | controller-manager | 10.211.55.22 |
4.2 生成kube-controller-manager證書
在運維主機200上操作。
生成證書請求檔案
/opt/certs/kube-controller-manager-csr.json
{
"CN": "system:kube-controller-manager",
"hosts": [
"127.0.0.1",
"10.211.55.11",
"10.211.55.12",
"10.211.55.21",
"10.211.55.22",
"10.211.55.23"
],
"key": {
"algo": "rsa",
"size": 2048
},
"names": [
{
"C": "CN",
"ST": "beijing",
"L": "beijing",
"O": "system:masters",
"OU": "system"
}
]
}
說明:
- CN:這裡的CN值非常重要,kube-controller-manager能否正常與kubee-apiserver通訊與此值有關,K8S預設會提取CN欄位的值作為使用者名稱,這實際是指K8S的“RoleBinding/ClusterRoleBinding”資源中“subjects:kind”的值為“User”
- hosts:kube-controller-manager執行節點的IP地址。
- O:無實際意義。
- OU:無實際意義。
- hosts 列表包含所有 kube-controller-manager 節點 IP;
- CN 為 system:kube-controller-manager、O 為 system:kube-controller-manager,kubernetes 內建的 ClusterRoleBindings system:kube-controller-manager 賦予 kube-controller-manager 工作所需的許可權
生成證書
[root@cfzx55-200 certs]# cfssl gencert \
> -ca=ca.pem \
> -ca-key=ca-key.pem \
> -config=ca-config.json \
> -profile=kubernetes \
> kube-controller-manager-csr.json | cfssl-json -bare kube-controller-manager
2022/03/13 10:35:55 [INFO] generate received request
2022/03/13 10:35:55 [INFO] received CSR
2022/03/13 10:35:55 [INFO] generating key: rsa-2048
2022/03/13 10:35:55 [INFO] encoded CSR
2022/03/13 10:35:55 [INFO] signed certificate with serial number 386505557530275475753178134460007976778023939766
[root@cfzx55-200 certs]# ll kube-controller*.pem
-rw------- 1 root root 1679 Mar 13 10:35 kube-controller-manager-key.pem
-rw-r--r-- 1 root root 1501 Mar 13 10:35 kube-controller-manager.pem
[root@cfzx55-200 certs]#
把證書拷貝到21和22主機上
[root@cfzx55-200 certs]# scp kube-controller-manager*.pem root@cfzx55-21:/opt/kubernetes/bin/certs/
[root@cfzx55-200 certs]# scp kube-controller-manager*.pem root@cfzx55-22:/opt/kubernetes/bin/certs/
4.3 生成kube-controller-manager的kubeconfig配置檔案
配置檔案路徑:/opt/kubernetes/cfg/
編寫生成kubeconfig配置檔案的指令碼
#!/bin/bash
KUBE_CONFIG="/opt/kubernetes/cfg/kube-controller-manager.kubeconfig"
KUBE_APISERVER="https://10.211.55.10:7443"
kubectl config set-cluster kubernetes \
--certificate-authority=/opt/kubernetes/bin/certs/ca.pem \
--embed-certs=true \
--server=${KUBE_APISERVER} \
--kubeconfig=${KUBE_CONFIG}
kubectl config set-credentials kube-controller-manager \
--client-certificate=/opt/kubernetes/bin/certs/kube-controller-manager.pem \
--client-key=/opt/kubernetes/bin/certs/kube-controller-manager-key.pem \
--embed-certs=true \
--kubeconfig=${KUBE_CONFIG}
kubectl config set-context default \
--cluster=kubernetes \
--user=kube-controller-manager \
--kubeconfig=${KUBE_CONFIG}
kubectl config use-context default --kubeconfig=${KUBE_CONFIG}
生成配置檔案
[root@cfzx55-21 k8s-shell]# vim kube-controller-manager-config.sh
[root@cfzx55-21 k8s-shell]# chmod +x kube-controller-manager-config.sh
[root@cfzx55-21 k8s-shell]# ./kube-controller-manager-config.sh
Cluster "kubernetes" set.
User "kube-controller-manager" set.
Context "default" created.
Switched to context "default".
[root@cfzx55-21 k8s-shell]#
把生成的配置檔案拷貝到22主機上。
[root@cfzx55-21 ~]# scp -r /opt/kubernetes/cfg/ root@cfzx55-22:/opt/kubernetes/
root@cfzx55-22's password:
kube-controller-manager.kubeconfig 100% 6366 2.6MB/s 00:00
[root@cfzx55-21 ~]#
在22主機上檢視
[root@cfzx55-22 ~]# ll /opt/kubernetes/cfg/
total 8
-rw------- 1 root root 6366 Mar 13 10:49 kube-controller-manager.kubeconfig
[root@cfzx55-22 ~]#
4.4 建立啟動指令碼
在21主機上操作
/opt/kubernetes/bin/kube-controller-manager-startup.sh
#!/bin/sh
./kube-controller-manager \
--cluster-name=kubernetes \
--bind-address=127.0.0.1 \
--service-cluster-ip-range=192.168.0.0/16 \
--leader-elect=true \
--controllers=*,bootstrapsigner,tokencleaner \
--kubeconfig=/opt/kubernetes/cfg/kube-controller-manager.kubeconfig \
--tls-cert-file=./certs/kube-controller-manager.pem \
--tls-private-key-file=./certs/kube-controller-manager-key.pem \
--cluster-signing-cert-file=./certs/ca.pem \
--cluster-signing-key-file=./certs/ca-key.pem \
--cluster-signing-duration=175200h0m0s \
--use-service-account-credentials=true \
--root-ca-file=./certs/ca.pem \
--service-account-private-key-file=./certs/ca-key.pem \
--log-dir=/data/logs/kubernetes/kube-controller-manager \
--v=2
說明:
--secure-port=10252 這個引數去掉,狀態才能正常。
--cluster-cidr string
CIDR Range for Pods in cluster. Requires --allocate-node-cidrs to be true
本例中,allocate-node-cidrs和cluster-cidr兩個引數不配置,使用docker的bip。
建立指令碼,調整許可權
[root@cfzx55-21 bin]# vim kube-controller-manager-startup.sh
[root@cfzx55-21 bin]# chmod +x kube-controller-manager-startup.sh
[root@cfzx55-21 bin]#
建立目錄
[root@cfzx55-21 ~]# mkdir -p /data/logs/kubernetes/kube-controller-manager
4.5 建立supervisor配置檔案
/etc/supervisord.d/kube-controller-manager.ini
[program:kube-controller-manager-55-21]
command=/opt/kubernetes/bin/kube-controller-manager-startup.sh
numprocs=1
directory=/opt/kubernetes/bin
autostart=true
autorestart=true
startsecs=30
startretries=3
exitcodes=0,2
stopsignal=QUIT
stopwaitsecs=10
user=root
redirect_stderr=true
stdout_logfile=/data/logs/kubernetes/kube-controller-manager/controller.stdout.log
stdout_logfile_maxbytes=64MB
stdout_logfile_backups=4
stdout_capture_maxbytes=1MB
stdout_events_enabled=false
4.6 啟動supervisor
[root@cfzx55-21 bin]# supervisorctl start kube-controller-manager-55-21
kube-controller-manager-55-21: started
[root@cfzx55-21 bin]# supervisorctl status
etcd-server-55-21 RUNNING pid 1033, uptime 4:21:51
kube-apiserver-55-21 RUNNING pid 1034, uptime 4:21:51
kube-controller-manager-55-21 RUNNING pid 3330, uptime 0:00:37
[root@cfzx55-21 bin]#
[root@cfzx55-21 bin]# netstat -luntp | grep kube
tcp 0 0 0.0.0.0:6443 0.0.0.0:* LISTEN 1044/./kube-apiserv
tcp 0 0 127.0.0.1:10252 0.0.0.0:* LISTEN 3331/./kube-control
[root@cfzx55-21 bin]#
4.7 把啟動指令碼、supervisor配置檔案拷貝到22主機。
[root@cfzx55-21 bin]# scp kube-controller-manager-startup.sh root@cfzx55-22:/opt/kubernetes/bin/
root@cfzx55-22's password:
kube-controller-manager-startup.sh 100% 778 489.1KB/s 00:00
[root@cfzx55-21 bin]# scp /etc/supervisord.d/kube-controller-manager.ini root@cfzx55-22:/etc/supervisord.d/
root@cfzx55-22's password:
kube-controller-manager.ini 100% 474 326.8KB/s 00:00
[root@cfzx55-21 bin]#
4.8 在22主機上啟動服務
# 修改程式名稱
[root@cfzx55-22 ~]# vim /etc/supervisord.d/kube-controller-manager.ini
[root@cfzx55-22 ~]# mkdir -p /data/logs/kubernetes/kube-controller-manager
[root@cfzx55-22 ~]# supervisorctl update
kube-controller-manager-55-21: added process group
[root@cfzx55-22 ~]# supervisorctl status
etcd-server-55-22 RUNNING pid 1013, uptime 4:27:39
kube-apiserver-55-22 RUNNING pid 1012, uptime 4:27:39
kube-controller-manager-55-21 RUNNING pid 3099, uptime 0:00:34
[root@cfzx55-22 ~]# netstat -luntp | grep kube
tcp 0 0 0.0.0.0:6443 0.0.0.0:* LISTEN 1014/./kube-apiserv
tcp 0 0 127.0.0.1:10252 0.0.0.0:* LISTEN 3100/./kube-control
[root@cfzx55-22 ~]#
5 部署kube-scheduler
5.1 叢集規劃
主機名 | 角色 | IP |
---|---|---|
CFZX55-21.host.com | kube-scheduler | 10.211.55.21 |
CFZX55-22.host.com | kube-scheduler | 10.211.55.22 |
5.2 生成kube-scheduler證書
建立證書請求csr檔案
/opt/certs/kube-scheduler-csr.json
{
"CN": "system:kube-scheduler",
"hosts": [
"127.0.0.1",
"10.211.55.11",
"10.211.55.12",
"10.211.55.21",
"10.211.55.22",
"10.211.55.23"
],
"key": {
"algo": "rsa",
"size": 2048
},
"names": [
{
"C": "CN",
"ST": "beijing",
"L": "beijing",
"O": "system:masters",
"OU": "system"
}
]
}
生成證書
[root@cfzx55-200 certs]# cfssl gencert \
> -ca=ca.pem \
> -ca-key=ca-key.pem \
> -config=ca-config.json \
> -profile=kubernetes \
> kube-scheduler-csr.json | cfssl-json -bare kube-scheduler
2022/03/13 12:30:21 [INFO] generate received request
2022/03/13 12:30:21 [INFO] received CSR
2022/03/13 12:30:21 [INFO] generating key: rsa-2048
2022/03/13 12:30:21 [INFO] encoded CSR
2022/03/13 12:30:21 [INFO] signed certificate with serial number 78101929142938232987965103781662806513424359272
[root@cfzx55-200 certs]# ll kube-scheduler*.pem
-rw------- 1 root root 1679 Mar 13 12:30 kube-scheduler-key.pem
-rw-r--r-- 1 root root 1489 Mar 13 12:30 kube-scheduler.pem
[root@cfzx55-200 certs]#
5.3 把證書拷貝到21和22節點。
[root@cfzx55-200 certs]# scp kube-scheduler*.pem root@cfzx55-21:/opt/kubernetes/bin/certs/
root@cfzx55-21's password:
kube-scheduler-key.pem 100% 1679 957.6KB/s 00:00
kube-scheduler.pem 100% 1489 953.3KB/s 00:00
[root@cfzx55-200 certs]# scp kube-scheduler*.pem root@cfzx55-22:/opt/kubernetes/bin/certs/
root@cfzx55-22's password:
kube-scheduler-key.pem 100% 1679 640.6KB/s 00:00
kube-scheduler.pem 100% 1489 794.6KB/s 00:00
[root@cfzx55-200 certs]#
5.4 生成kubeconfig配置檔案
#!/bin/bash
KUBE_CONFIG="/opt/kubernetes/cfg/kube-scheduler.kubeconfig"
KUBE_APISERVER="https://10.211.55.10:7443"
kubectl config set-cluster kubernetes \
--certificate-authority=/opt/kubernetes/bin/certs/ca.pem \
--embed-certs=true \
--server=${KUBE_APISERVER} \
--kubeconfig=${KUBE_CONFIG}
kubectl config set-credentials kube-scheduler \
--client-certificate=/opt/kubernetes/bin/certs/kube-scheduler.pem \
--client-key=/opt/kubernetes/bin/certs/kube-scheduler-key.pem \
--embed-certs=true \
--kubeconfig=${KUBE_CONFIG}
kubectl config set-context default \
--cluster=kubernetes \
--user=kube-scheduler \
--kubeconfig=${KUBE_CONFIG}
kubectl config use-context default --kubeconfig=${KUBE_CONFIG}
執行指令碼
[root@cfzx55-21 k8s-shell]# vim kube-scheduler-config.sh
[root@cfzx55-21 k8s-shell]# chmod +x kube-scheduler-config.sh
[root@cfzx55-21 k8s-shell]# ./kube-scheduler-config.sh
Cluster "kubernetes" set.
User "kube-scheduler" set.
Context "default" created.
Switched to context "default".
[root@cfzx55-21 k8s-shell]#
把kubeconfig 檔案拷貝到22主機
[root@cfzx55-21 k8s-shell]# scp /opt/kubernetes/cfg/kube-scheduler.kubeconfig root@cfzx55-22:/opt/kubernetes/cfg/
root@cfzx55-22's password:
kube-scheduler.kubeconfig 100% 6332 2.6MB/s 00:00
[root@cfzx55-21 k8s-shell]#
5.5 建立kube-scheduler啟動指令碼
/opt/kubernetes/bin/kube-scheduler-startup.sh
#!/bin/sh
./kube-scheduler \
--address=127.0.0.1 \
--leader-elect=true \
--kubeconfig=/opt/kubernetes/cfg/kube-scheduler.kubeconfig \
--log-dir=/data/logs/kubernetes/kube-scheduler \
--v=2
建立指令碼,調整許可權
[root@cfzx55-21 bin]# vim kube-scheduler-startup.sh
[root@cfzx55-21 bin]# chmod +x kube-scheduler-startup.sh
[root@cfzx55-21 ~]# mkdir -p /data/logs/kubernetes/kube-scheduler
5.6 建立supervisor配置檔案
/etc/supervisord.d/kube-scheduler.ini
[program:kube-scheduler-55-21]
command=/opt/kubernetes/bin/kube-scheduler-startup.sh
numprocs=1
directory=/opt/kubernetes/bin
autostart=true
autorestart=true
startsecs=30
startretries=3
exitcodes=0,2
stopsignal=QUIT
stopwaitsecs=10
user=root
redirect_stderr=true
stdout_logfile=/data/logs/kubernetes/kube-scheduler/scheduler.stdout.log
stdout_logfile_maxbytes=64MB
stdout_logfile_backups=4
stdout_capture_maxbytes=1MB
stdout_events_enabled=false
5.7 啟動kube-scheduler服務
[root@cfzx55-21 bin]# supervisorctl update
[root@cfzx55-21 bin]# supervisorctl status
etcd-server-55-21 RUNNING pid 1033, uptime 5:16:26
kube-apiserver-55-21 RUNNING pid 1034, uptime 5:16:26
kube-controller-manager-55-21 RUNNING pid 3416, uptime 0:38:46
kube-scheduler-55-21 RUNNING pid 3486, uptime 0:00:32
[root@cfzx55-21 bin]# netstat -luntp | grep kube
tcp 0 0 0.0.0.0:10259 0.0.0.0:* LISTEN 3487/./kube-schedul
tcp 0 0 0.0.0.0:6443 0.0.0.0:* LISTEN 1044/./kube-apiserv
tcp 0 0 127.0.0.1:10252 0.0.0.0:* LISTEN 3417/./kube-control
[root@cfzx55-21 bin]#
5.8 把kube-scheduler啟動指令碼和supervisor配置檔案拷貝到22主機
[root@cfzx55-21 bin]# scp kube-scheduler-startup.sh root@cfzx55-22:/opt/kubernetes/bin/
root@cfzx55-22's password:
kube-scheduler-startup.sh 100% 199 100.3KB/s 00:00
[root@cfzx55-21 bin]#
[root@cfzx55-21 bin]# scp /etc/supervisord.d/kube-scheduler.ini root@cfzx55-22:/etc/supervisord.d/
root@cfzx55-22's password:
kube-scheduler.ini 100% 446 329.4KB/s 00:00
[root@cfzx55-21 bin]#
5.9 在22主機上啟動kube-scheduler服務
# 修改名稱
[root@cfzx55-22 ~]# vim /etc/supervisord.d/kube-scheduler.ini
[root@cfzx55-22 ~]# supervisorctl update
kube-controller-manager-55-21: stopped
kube-controller-manager-55-21: removed process group
kube-controller-manager-55-22: added process group
kube-scheduler-55-22: added process group
[root@cfzx55-22 ~]#
[root@cfzx55-22 ~]# supervisorctl status
etcd-server-55-22 RUNNING pid 1013, uptime 5:25:59
kube-apiserver-55-22 RUNNING pid 1012, uptime 5:25:59
kube-controller-manager-55-22 RUNNING pid 3234, uptime 0:00:32
kube-scheduler-55-22 RUNNING pid 3187, uptime 0:03:19
[root@cfzx55-22 ~]# netstat -luntp | grep kube
tcp 0 0 0.0.0.0:6443 0.0.0.0:* LISTEN 1014/./kube-apiserv
tcp 0 0 127.0.0.1:10252 0.0.0.0:* LISTEN 3235/./kube-control
tcp 0 0 0.0.0.0:10259 0.0.0.0:* LISTEN 3189/./kube-schedul
[root@cfzx55-22 ~]#
部署22上的kube-controller-manager時,沒有修改名稱。
檢視叢集狀態
[root@cfzx55-22 bin]# kubectl get cs
Warning: v1 ComponentStatus is deprecated in v1.19+
NAME STATUS MESSAGE ERROR
controller-manager Healthy ok
scheduler Healthy ok
etcd-2 Healthy {"health":"true","reason":""}
etcd-0 Healthy {"health":"true","reason":""}
etcd-1 Healthy {"health":"true","reason":""}
[root@cfzx55-22 bin]#
檢視叢集資源
[root@cfzx55-21 ~]# kubectl get sa -A
NAMESPACE NAME SECRETS AGE
default default 1 114m
kube-node-lease default 1 114m
kube-public default 1 114m
kube-system attachdetach-controller 1 116m
kube-system bootstrap-signer 1 114m
kube-system certificate-controller 1 116m
kube-system clusterrole-aggregation-controller 1 114m
kube-system cronjob-controller 1 116m
kube-system daemon-set-controller 1 116m
kube-system default 1 114m
kube-system deployment-controller 1 116m
kube-system disruption-controller 1 116m
kube-system endpoint-controller 1 114m
kube-system endpointslice-controller 1 116m
kube-system endpointslicemirroring-controller 1 116m
kube-system ephemeral-volume-controller 1 116m
kube-system expand-controller 1 114m
kube-system generic-garbage-collector 1 114m
kube-system horizontal-pod-autoscaler 1 116m
kube-system job-controller 1 116m
kube-system namespace-controller 1 116m
kube-system node-controller 1 116m
kube-system persistent-volume-binder 1 114m
kube-system pod-garbage-collector 1 114m
kube-system pv-protection-controller 1 114m
kube-system pvc-protection-controller 1 114m
kube-system replicaset-controller 1 116m
kube-system replication-controller 1 114m
kube-system resourcequota-controller 1 114m
kube-system root-ca-cert-publisher 1 116m
kube-system service-account-controller 1 116m
kube-system service-controller 1 116m
kube-system statefulset-controller 1 116m
kube-system token-cleaner 1 114m
kube-system ttl-after-finished-controller 1 114m
kube-system ttl-controller 1 116m
[root@cfzx55-21 ~]# kubectl get ns -A
NAME STATUS AGE
default Active 15h
kube-node-lease Active 15h
kube-public Active 15h
kube-system Active 15h
[root@cfzx55-21 ~]# kubectl get role -A
NAMESPACE NAME CREATED AT
kube-public system:controller:bootstrap-signer 2022-03-12T14:36:17Z
kube-system extension-apiserver-authentication-reader 2022-03-12T14:36:16Z
kube-system system::leader-locking-kube-controller-manager 2022-03-12T14:36:16Z
kube-system system::leader-locking-kube-scheduler 2022-03-12T14:36:16Z
kube-system system:controller:bootstrap-signer 2022-03-12T14:36:16Z
kube-system system:controller:cloud-provider 2022-03-12T14:36:16Z
kube-system system:controller:token-cleaner 2022-03-12T14:36:16Z
[root@cfzx55-21 ~]#
至此,Master節點部署完成。