二进制安装-k8s高可用集群07-部署高可用组件
本文档讲解使用 keepalived 和 haproxy 实现 kube-apiserver 高可用的步骤:
- keepalived 提供 kube-apiserver 对外服务的 VIP;
- haproxy 监听 VIP,后端连接所有 kube-apiserver 实例,提供健康检查和负载均衡功能;
运行 keepalived 和 haproxy 的节点称为 LB 节点。由于 keepalived 是一主多备运行模式,故至少两个 LB 节点。
本文档复用 master 节点的三台机器,haproxy 监听的端口 (8443) 需要与 kube-apiserver 的端口 6443 不同,避免冲突。
keepalived 在运行过程中周期检查本机的 haproxy 进程状态,如果检测到 haproxy 进程异常,则触发重新选主的过程,VIP 将飘移到新选出来的主节点,从而实现 VIP 的高可用。
所有组件(如 kubeclt、apiserver、controller-manager、scheduler 等)都通过 VIP 和 haproxy 监听的 8443 端口访问 kube-apiserver 服务。
1)安装软件包
cat > magic27_install_package.sh << "EOF"
#!/bin/bash
# 安装软件包
source /opt/k8s/bin/environment.sh
for node_ip in ${NODE_IPS[@]}
do
echo ">>> ${node_ip}"
ssh root@${node_ip} "yum install -y keepalived haproxy"
done
EOF
2)配置和下发 haproxy 配置文件
haproxy 配置文件:
cat > haproxy.cfg <<EOF
global
log /dev/log local0
log /dev/log local1 notice
chroot /var/lib/haproxy
stats socket /var/run/haproxy-admin.sock mode 660 level admin
stats timeout 30s
user haproxy
group haproxy
daemon
nbproc 1
defaults
log global
timeout connect 5000
timeout client 10m
timeout server 10m
listen admin_stats
bind 0.0.0.0:10080
mode http
log 127.0.0.1 local0 err
stats refresh 30s
stats uri /status
stats realm welcome login\ Haproxy
stats auth admin:123456
stats hide-version
stats admin if TRUE
listen kube-master
bind 0.0.0.0:8443
mode tcp
option tcplog
balance source
server 172.68.96.101 172.68.96.101:6443 check inter 2000 fall 2 rise 2 weight 1
server 172.68.96.102 172.68.96.102:6443 check inter 2000 fall 2 rise 2 weight 1
server 172.68.96.103 172.68.96.103:6443 check inter 2000 fall 2 rise 2 weight 1
EOF
注意更改其中的 ip 与自己的一致。
- haproxy 在 10080 端口输出 status 信息;
- haproxy 监听所有接口的 8443 端口,该端口与环境变量 ${KUBE_APISERVER} 指定的端口必须一致;
- server 字段列出所有 kube-apiserver 监听的 IP 和端口;
分发 haproxy.cfg 到所有集群节点上:
cat > magic28_distribute_haproxy.sh << "EOF"
#!/bin/bash
# 分发 haproxy.cfg 到所有集群节点上
source /opt/k8s/bin/environment.sh
for node_ip in ${NODE_IPS[@]}
do
echo ">>> ${node_ip}"
scp /data/template/haproxy.cfg root@${node_ip}:/etc/haproxy
done
EOF
3)启动 haproxy 服务
cat > magic29_start_haproxy.sh << "EOF"
#!/bin/bash
# 启动 haproxy 服务
source /opt/k8s/bin/environment.sh
for node_ip in ${NODE_IPS[@]}
do
echo ">>> ${node_ip}"
ssh root@${node_ip} "systemctl restart haproxy"
done
EOF
4)检查 haproxy 服务状态
cat > magic30_check_haproxy_service.sh << "EOF"
#!/bin/bash
# 检查 haproxy 服务状态
source /opt/k8s/bin/environment.sh
for node_ip in ${NODE_IPS[@]}
do
echo ">>> ${node_ip}"
ssh root@${node_ip} "systemctl status haproxy|grep Active"
done
EOF
如果看到如下输出:
bash magic30_check_haproxy_service.sh
>>> 172.68.96.101
Active: active (running) since Wed 20XX-XX-XX XX:XX:XX CST; XXh ago
>>> 172.68.96.102
Active: active (running) since Wed 20XX-XX-XX XX:XX:XX CST; XXh ago
>>> 172.68.96.103
Active: active (running) since Wed 20XX-XX-XX XX:XX:XX CST; XXh ago
则说明正常,如果失败,用如下命令检查:
journalctl -xu haproxy
检查 haproxy 是否监听 8443 端口:
cat > magic31_check_haproxy_proxy8443.sh << "EOF"
#!/bin/bash
source /opt/k8s/bin/environment.sh
for node_ip in ${NODE_IPS[@]}
do
echo ">>> ${node_ip}"
ssh root@${node_ip} "netstat -lnpt|grep haproxy"
done
EOF
如果看到如下输出:
bash magic31_check_haproxy_proxy8443.sh
>>> 172.68.96.101
tcp 0 0 0.0.0.0:8443 0.0.0.0:* LISTEN 25950/haproxy
tcp 0 0 0.0.0.0:10080 0.0.0.0:* LISTEN 25950/haproxy
>>> 172.68.96.102
tcp 0 0 0.0.0.0:8443 0.0.0.0:* LISTEN 8042/haproxy
tcp 0 0 0.0.0.0:10080 0.0.0.0:* LISTEN 8042/haproxy
>>> 172.68.96.103
tcp 0 0 0.0.0.0:8443 0.0.0.0:* LISTEN 7391/haproxy
tcp 0 0 0.0.0.0:10080 0.0.0.0:* LISTEN 7391/haproxy
5)配置和下发 keepalived 配置文件
keepalived 是一主(master)多备(backup)运行模式,故有两种类型的配置文件。master 配置文件只有一份,backup 配置文件视节点数目而定,对于本文档而言,规划如下:
- master: 172.68.96.101
- backup:172.68.96.102,172.68.96.103
master 配置文件:
source /opt/k8s/bin/environment.sh
cat > /data/template/keepalived-master.conf <<EOF
global_defs {
router_id lb-master-105
}
vrrp_script check-haproxy {
script "killall -0 haproxy"
interval 5
weight -30
}
vrrp_instance VI-kube-master {
state MASTER
priority 120
dont_track_primary
interface ${VIP_IF}
virtual_router_id 68
advert_int 3
track_script {
check-haproxy
}
virtual_ipaddress {
${MASTER_VIP}
}
}
EOF
- VIP 所在的接口(interface ${VIP_IF})为 eth0;
- 使用 killall -0 haproxy 命令检查所在节点的 haproxy 进程是否正常。如果异常则将权重减少(-30), 从而触发重新选主过程;
- router_id、virtual_router_id 用于标识属于该 HA 的 keepalived 实例,如果有多套 keepalived HA,则必须各不相同;
backup 配置文件:
source /opt/k8s/bin/environment.sh
cat > /data/template/keepalived-backup.conf <<EOF
global_defs {
router_id lb-backup-105
}
vrrp_script check-haproxy {
script "killall -0 haproxy"
interval 5
weight -30
}
vrrp_instance VI-kube-master {
state BACKUP
priority 110
dont_track_primary
interface ${VIP_IF}
virtual_router_id 68
advert_int 3
track_script {
check-haproxy
}
virtual_ipaddress {
${MASTER_VIP}
}
}
EOF
- VIP 所在的接口(interface ${VIP_IF})为 eth0;
- 使用 killall -0 haproxy 命令检查所在节点的 haproxy 进程是否正常。如果异常则将权重减少(-30), 从而触发重新选主过程;
- router_id、virtual_router_id 用于标识属于该 HA 的 keepalived 实例,如果有多套 keepalived HA,则必须各不相同;
- priority 的值必须小于 master 的值;
6)分发 keepalived 配置文件
下发 master 配置文件:
scp /data/template/keepalived-master.conf root@master:/etc/keepalived/keepalived.conf
下发 backup 配置文件:
scp /data/template/keepalived-backup.conf root@node01:/etc/keepalived/keepalived.conf
scp /data/template/keepalived-backup.conf root@node02:/etc/keepalived/keepalived.conf
7)启动 keepalived 服务
cat > magic32_start_keepalived_service.sh << "EOF"
#!/bin/bash
# 启动 keepalived 服务
source /opt/k8s/bin/environment.sh
for node_ip in ${NODE_IPS[@]}
do
echo ">>> ${node_ip}"
ssh root@${node_ip} "systemctl restart keepalived"
done
EOF
8)检查 keepalived 服务
cat > magic33_check_keepalived_service.sh << "EOF"
#!/bin/bash
# 检查 keepalived 服务
source /opt/k8s/bin/environment.sh
for node_ip in ${NODE_IPS[@]}
do
echo ">>> ${node_ip}"
ssh root@${node_ip} "systemctl status keepalived|grep Active"
done
EOF
输出如下:
bash magic33_check_keepalived_service.sh
>>> 172.68.96.101
Active: active (running) since Wed 20XX-XX-XX XX:XX:XX CST; XXh ago
>>> 172.68.96.102
Active: active (running) since Wed 20XX-XX-XX XX:XX:XX CST; XXh ago
>>> 172.68.96.103
Active: active (running) since Wed 20XX-XX-XX XX:XX:XX CST; XXh ago
则正常,如果失败,则检查日志
journalctl -xu keepalived
查看 VIP 所在的节点,确保可以 ping 通 VIP:
cat > magic34_ping_keepalived_service.sh << "EOF"
#!/bin/bash
# 查看VIP所在的节点,确保可以ping通VIP
source /opt/k8s/bin/environment.sh
for node_ip in ${NODE_IPS[@]}
do
echo ">>> ${node_ip}"
ssh ${node_ip} "/usr/sbin/ip addr show ${VIP_IF}"
ssh ${node_ip} "ping -c 1 ${MASTER_VIP}"
done
EOF
可以看到 VIP 目前在 master 节点上,而且各个节点也都是通的
bash magic34_ping_keepalived_service.sh
>>> 172.68.96.101
......
PING 172.68.96.88 (172.68.96.88) 56(84) bytes of data.
64 bytes from 172.68.96.88: icmp_seq=1 ttl=64 time=0.031 ms
...........
9)查看 haproxy 状态页面
浏览器访问 ${MASTER_VIP}:10080/status 地址,查看 haproxy 状态页面:
用户名密码就在刚刚定义的 haproxy 的配置当中。