企业级Keepalived高可用离线部署实战（附K8S集群VIP配置）

news2025/5/9 18:27:11

企业级Keepalived高可用离线部署实战（附K8S集群VIP配置）

摘要：本文详细讲解在离线环境下部署Keepalived实现Kubernetes集群高可用的完整流程，涵盖源码编译安装、多节点配置、健康检查联动等核心环节，并提供生产级参数调优方案。适用于金融、政务等隔离网络的容灾架构建设。

一、环境规划与架构设计

1.1 节点信息

节点IP	角色	优先级	网卡接口
192.167.14.119	MASTER	101	eno3
192.167.14.223	BACKUP	100	ens192
192.167.14.226	BACKUP	99	ens192

1.2 虚拟IP（VIP）

VIP地址：192.167.14.205
作用：作为Kubernetes API Server的统一接入点，实现故障自动切换

二、离线安装Keepalived

2.1 源码编译安装

# 下载源码包（需提前在有网环境操作）
wget https://www.keepalived.org/software/keepalived-2.2.8.tar.gz

# 解压并编译
tar xvf keepalived-2.2.8.tar.gz
cd keepalived-2.2.8
./configure --prefix=/usr/local/keepalived --disable-track-process
make && make install

2.2 系统服务配置

# 复制配置文件
cp /usr/local/keepalived/etc/sysconfig/keepalived /etc/sysconfig/
cp /usr/local/keepalived/sbin/keepalived /usr/sbin/

# 配置启动脚本
cp keepalived-2.2.8/keepalived/etc/init.d/keepalived /etc/init.d/
chmod +x /etc/init.d/keepalived

# 创建配置目录
mkdir /etc/keepalived
cp /usr/local/keepalived/etc/keepalived/keepalived.conf.sample /etc/keepalived/keepalived.conf

# 设置开机启动
chkconfig --add keepalived
chkconfig keepalived on

三、多节点配置详解

3.1 MASTER节点配置（119）

cat > /etc/keepalived/keepalived.conf << EOF
! Configuration File for keepalived
global_defs {
    router_id K8S_MASTER_119  # 唯一标识
    enable_script_security    # 启用脚本安全
}

vrrp_script chk_apiserver {
    script "/etc/keepalived/check_apiserver.sh"  # 健康检测脚本
    interval 5     # 检测间隔(秒)
    weight -5      # 检测失败权重变化
    fall 2         # 连续失败2次判定异常
    rise 1         # 成功1次恢复
}

vrrp_instance VI_1 {
    state MASTER            # 初始状态
    interface eno3          # 绑定网卡
    virtual_router_id 51    # 集群ID(同一组需相同)
    priority 101            # 优先级(MASTER最高)
    advert_int 2            # 心跳间隔
    
    authentication {        # 认证配置
        auth_type PASS
        auth_pass K8SHA_KA_AUTH
    }
    
    virtual_ipaddress {     # 虚拟IP配置
        192.167.14.205
    }
    
    track_script {         # 关联检测脚本
        chk_apiserver
    }
}
EOF

3.2 BACKUP节点配置（223/226）

# 以223节点为例，修改以下参数：
state BACKUP
interface ens192
mcast_src_ip 192.167.14.223
priority 100  # 226节点设为99

四、健康检查联动机制

4.1 检测脚本配置

cat > /etc/keepalived/check_apiserver.sh << EOF
#!/bin/bash
timeout=3
err=0

# 检测haproxy服务状态
for ((i=0; i<timeout; i++)); do
    if systemctl is-active --quiet haproxy; then
        break
    else
        err=$((err+1))
        sleep 1
    fi
done

# 服务异常时停止keepalived触发切换
if [[ $err -eq $timeout ]]; then
    systemctl stop keepalived
    exit 1
else
    exit 0
fi
EOF

chmod +x /etc/keepalived/check_apiserver.sh

4.2 服务状态联动逻辑

五、服务启动与验证

5.1 启动服务

# 所有节点执行
systemctl daemon-reload
systemctl enable --now keepalived
systemctl status keepalived

5.2 VIP验证命令

# 查看VIP绑定情况
ip addr show eno3 | grep 192.167.14.205

# 模拟故障测试
MASTER节点执行：
systemctl stop haproxy
tail -f /var/log/messages  # 观察切换日志

六、生产环境调优建议

6.1 安全加固措施

# 限制VRRP协议通信(所有节点)
iptables -A INPUT -p vrrp -s 192.167.14.0/24 -j ACCEPT
iptables -A INPUT -p vrrp -j DROP

6.2 参数优化

vrrp_instance VI_1 {
    ...
    nopreempt         # 禁止抢占，避免网络抖动
    debug 1           # 调试日志级别
    notify_master "/path/to/notify.sh master"  # 状态切换通知
}

七、常见问题排查指南

故障现象	排查命令	解决方案
VIP未正确绑定	`ip addr show`	检查网卡配置、防火墙规则
节点状态始终为BACKUP	`journalctl -u keepalived`	确认priority配置高于其他节点
健康检查误触发切换	`systemctl status haproxy`	调整检测间隔/失败阈值
节点间无法通信	`tcpdump -i eno3 vrrp`	检查网络连通性、组播配置