结合使用，实现IPVS的高可用性、利用VRRP Script 实现全能高可用

脆皮的饭桶2026-03-22 19:13

一、实现IPVS的高可用性

1.1 简介

1.1.1 IPVS 高可用核心方案：IPVS + Keepalived 双机热备

1.1.2 IPVS 的高可用本质是

用 Keepalived 做VIP 管理 + 故障切换，用 IPVS 做负载均衡转发，两者配合实现：

1.1.3 特点

流量始终可达：VIP 不随单节点故障失效

负载均衡不变：IPVS 规则在两台 LB 上一致

故障自动恢复：秒级切换，对业务无感知

二、 配置ipvs的高可用需要的环境

2.1 实验环境

|-------------|-----------------------------------------------------------|------|
| 主机名 | IP | 角色 |
| keepalived1 | 192.168.248.50 模式:仅主机 wg:192.168.248.2 | 调度器1 |
| keepalived2 | 192.168.248.60 模式:仅主机 wg:192.168.248.2 | 调度器2 |
| webserve1 | 192.168.248.10 模式:仅主机 wg:192.168.248.2 lo：192.168.248.100 | 服务器1 |
| webserver2 | 192.168.248.20 模式:仅主机 wg:192.168.248.2 lo：192.168.248.100 | 服务器2 |
| client | 192.168.198.10 模式:host-only | 客户端 |

2.2 配置webserver 的 ip，配置lvs相应的软件环境及文件

2.2.1 配置webserver IP lo回环ip

bash 复制代码 
[root@web1 system-connections]# cp eth0.nmconnection lo.nmconnection  -p
[root@web1 system-connections]# vim lo.nmconnection

[connection]
id=lo
type=loopback
interface-name=lo
[ipv4]
method=manual
address1=127.0.0.1/8
address1=192.168.248.100/32

[root@web1 system-connections]# nmcli connection reload
[root@web1 system-connections]# nmcli connection up lo
连接已成功激活（D-Bus 活动路径：/org/freedesktop/NetworkManager/ActiveConnection/5）

[root@web1 ~]#  ip ad
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    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
    inet 192.168.248.100/32 scope global lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether 00:0c:29:f2:dc:c4 brd ff:ff:ff:ff:ff:ff
    altname enp3s0
    altname ens160
    inet 192.168.248.10/24 brd 192.168.248.255 scope global noprefixroute eth0
       valid_lft forever preferred_lft forever
    inet6 fe80::a9f6:400:428:83d4/64 scope link noprefixroute
       valid_lft forever preferred_lft forever



[root@web2 system-connections]# cp eth0.nmconnection lo.nmconnection  -p
[root@web2 system-connections]# vim lo.nmconnection

[connection]
id=lo
type=loopback
interface-name=lo
[ipv4]
method=manual
address1=127.0.0.1/8
address1=192.168.248.100/32

[root@web2 system-connections]# nmcli connection reload
[root@web2 system-connections]# nmcli connection up lo
连接已成功激活（D-Bus 活动路径：/org/freedesktop/NetworkManager/ActiveConnection/5）


[root@web2 ~]# ip add
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    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
    inet 127.0.0.1/24 brd 127.0.0.255 scope host lo
       valid_lft forever preferred_lft forever
    inet 192.168.248.100/32 scope global lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether 00:0c:29:ad:be:2c brd ff:ff:ff:ff:ff:ff
    altname enp3s0
    altname ens160
    inet 192.168.248.20/24 brd 192.168.248.255 scope global noprefixroute eth0
       valid_lft forever preferred_lft forever
    inet6 fe80::8e01:e3ed:e8db:424a/64 scope link noprefixroute
       valid_lft forever preferred_lft forever

2.2.2 给webserver做ARP抑制配置

web 不做 ARP 抑制，它会对外广播我是 VIP，导致客户端流量直接打到 web ，绕过 keepalived ，破坏负载均衡逻辑

bash 复制代码 
[root@web1 ~]# vim /etc/sysctl.conf
[root@web1 ~]# sysctl -p
net.ipv4.conf.all.arp_ignore = 1
net.ipv4.conf.all.arp_announce = 2
net.ipv4.conf.lo.arp_ignore = 1
net.ipv4.conf.lo.arp_announce = 2


[root@web2 ~]# vim /etc/sysctl.conf
[root@web2 ~]# sysctl -p
net.ipv4.conf.all.arp_ignore = 1
net.ipv4.conf.all.arp_announce = 2
net.ipv4.conf.lo.arp_ignore = 1
net.ipv4.conf.lo.arp_announce = 2

net.ipv4.conf.all.arp_ignore=1 ：对所有网口**，**对所有网口，仅响应目标 IP 是本网卡 IP 的 ARP 请求，拒绝响应针对 VIP绑在 lo的 ARP 查询，避免 RS 抢 VIP

net.ipv4.conf.all.arp_announce=2：对所有网口，仅宣告本网卡的主 IP，不宣告次级 / 别名 IP 不对外广播我有 VIP，只暴露真实物理网卡 IP

net.ipv4.conf.lo.arp_announce=2：对 lo 接口，限制 ARP 宣告，不把 lo 上的 VIP 广播出去

net.ipv4.conf.lo.arp_ignore=1：对 lo 接口，仅响应发给 lo 自身 IP 的 ARP 请求，进一步隔离 VIP 的 ARP 广播

2.3 开启基础的ipvs高可用

下载 ipvsadm 配置keepalived.conf 文件

bash 复制代码 
[root@KA1 ~]# dnf install ipvsadm-1.31-6.el9.x86_64 -y
 17 vrrp_instance WEB_VIP {
 18     state MASTER
 19     interface eth0
 20     virtual_router_id 51
 21     priority 100
 22     advert_int 1
 23     authentication {
 24         auth_type PASS
 25         auth_pass 1111
 26     }
 27     virtual_ipaddress {
 28         192.168.248.100/24 dev eth0 label eth0:0
 29     }
 30 }

 33 virtual_server 192.168.248.100 80 {
 34     delay_loop 6
 35     lb_algo rr
 36     lb_kind DR
 37     protocol TCP
 38
 39 real_server 192.168.248.10 80 {
 40     weight 1
 41     HTTP_GET {
 42         url {
 43             path /
 44             status_code 200
 45             }
 46             connect_timeout 1
 47             retry 3
 48             delay_before_retry 1
 49         }
 50     }
 51 real_server 192.168.248.20 80 {
 52     weight 1
 53     TCP_CHECK {
 54         connect_timeout 5
 55         retry 3
 56         delay_before_retry 3
 57         connect_port 80
 58         }
 59     }
 60 }
[root@KA1 ~]# systemctl restart keepalived.service

[root@KA2 ~]# vim /etc/keepalived/keepalived.conf
 18 vrrp_instance WEB_VIP {
 19     state BACKUP
 20     interface eth0
 21     virtual_router_id 51
 22     priority 80
 23     advert_int 1
 24     authentication {
 25         auth_type PASS
 26         auth_pass 1111
 27     }
 28     virtual_ipaddress {
 29         192.168.248.100 dev eth0 label eth0:0
 30     }
 31 }
 32 virtual_server 192.168.248.100 80 {
 33     delay_loop 6
 34     lb_algo rr
 35     lb_kind DR
 36     protocol TCP
 37
 38     real_server 192.168.248.10 80 {
 39         weight 1
 40         HTTP_GET {
 41             url {
 42               path /
 43               status_code 200
 44             }
 45             connect_timeout 1
 46             retry 3
 47             delay_before_retry 1
 48       }
 49     }
 50
 51     real_server 192.168.248.20 80 {
 52         weight 1
 53         TCP_CHECK {
 54           connect_timeout 5
 55           retry 3
 56           delay_before_retry 3
 57           connect_port 80
 58       }
 59     }
 60 }
[root@KA2 ~]# systemctl restart keepalived.service

2.4 测试

查看策略是否自动生成

bash 复制代码 
[root@KA1 ~]# ipvsadm -Ln
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
  -> RemoteAddress:Port           Forward Weight ActiveConn InActConn
TCP  192.168.198.100:80 rr
  -> 192.168.248.10:80            Route   1      0          0
  -> 192.168.248.20:80            Route   1      0          0

curl 192.168.248.100 查看是否实现均衡负载

bash 复制代码 
[2026-03-21 19:22.38]  ~
[.LAPTOP-223ABRUC] ⮞ curl 192.168.248.100
RS1 - 192.168.248.10
                                                                                                   ✓
────────────────────────────────────────────────────────────────────────────────────────────────────
[2026-03-21 19:22.40]  ~
[.LAPTOP-223ABRUC] ⮞ curl 192.168.248.100
RS2 - 192.168.248.20
                                                                                                   ✓
────────────────────────────────────────────────────────────────────────────────────────────────────
[2026-03-21 19:22.41]  ~
[.LAPTOP-223ABRUC] ⮞ curl 192.168.248.100
RS1 - 192.168.248.10
                                                                                                   ✓
────────────────────────────────────────────────────────────────────────────────────────────────────
[2026-03-21 19:26.20]  ~
[.LAPTOP-223ABRUC] ⮞ curl 192.168.248.100
RS2 - 192.168.248.20

监控ipvsadm 策略编写，关闭KA1 的keepalived 服务，查看策略变化

bash 复制代码 
[root@KA1 ~]# systemctl stop keepalived.service
[root@KA1 ~]# ip add
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    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
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether 00:0c:29:6d:7f:2b brd ff:ff:ff:ff:ff:ff
    altname enp3s0
    altname ens160
    inet 192.168.248.50/24 brd 192.168.248.255 scope global noprefixroute eth0
       valid_lft forever preferred_lft forever
    inet6 fe80::7831:d38e:18e8:c266/64 scope link noprefixroute
       valid_lft forever preferred_lft forever
[root@KA1 ~]# ipvsadm -Ln
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
  -> RemoteAddress:Port           Forward Weight ActiveConn InActConn


[root@KA2 ~]# ip add
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    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
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether 00:0c:29:f6:15:27 brd ff:ff:ff:ff:ff:ff
    altname enp3s0
    altname ens160
    inet 192.168.248.60/24 brd 192.168.248.255 scope global noprefixroute eth0
       valid_lft forever preferred_lft forever
    inet 192.168.248.100/32 scope global eth0:0
       valid_lft forever preferred_lft forever
    inet6 fe80::3b5a:7484:2377:ed88/64 scope link noprefixroute
       valid_lft forever preferred_lft forever
[root@KA2 ~]# ipvsadm -Ln
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
  -> RemoteAddress:Port           Forward Weight ActiveConn InActConn
TCP  192.168.248.100:80 rr
  -> 192.168.248.10:80            Route   1      0          0
  -> 192.168.248.20:80            Route   1      0          0

测试 当vip 漂移到KA2 时是否还能实现均衡负载

bash 复制代码 
[2026-03-21 19:26.25]  ~
[.LAPTOP-223ABRUC] ⮞ curl 192.168.248.100
RS1 - 192.168.248.10
                                                                                                   ✓
────────────────────────────────────────────────────────────────────────────────────────────────────
[2026-03-21 19:30.46]  ~
[.LAPTOP-223ABRUC] ⮞ curl 192.168.248.100
RS2 - 192.168.248.20
                                                                                                   ✓
────────────────────────────────────────────────────────────────────────────────────────────────────
[2026-03-21 19:30.46]  ~
[.LAPTOP-223ABRUC] ⮞ curl 192.168.248.100
RS1 - 192.168.248.10
                                                                                                   ✓
────────────────────────────────────────────────────────────────────────────────────────────────────
[2026-03-21 19:30.47]  ~
[.LAPTOP-223ABRUC] ⮞ curl 192.168.248.100
RS2 - 192.168.248.20

三、双主模式代理不同业务实现高可用

3.1 简介：

Keepalived + LVS的进阶高可用方案： 两台节点同时对外提供服务，各自承载不同业务，同时互为备份，解决了主备模式下备机闲置的问题，资源利用率更高。

优势：

（1）资源利用率高：两台机器都在干活，避免备机空转

（2）业务隔离：不同业务用不同 VIP，故障影响范围小

（3）高可用更强：单节点故障不影响任何业务，自动漂移

（4）水平扩展：后续可加更多节点，扩展更多业务

3.2 双主模式代理http 服务和 mysql 服务

在web1 web2 上下载 数据库 maridb，做初始化配置

bash 复制代码 
[root@web1 ~]# dnf install mariadb-server -y
systemctl enable --now mariadb
[root@web1 ~]# systemctl enable --now mariadb
Created symlink /etc/systemd/system/mysql.service → /usr/lib/systemd/system/mariadb.se                                                                    rvice.
Created symlink /etc/systemd/system/mysqld.service → /usr/lib/systemd/system/mariadb.s                                                                    ervice.
Created symlink /etc/systemd/system/multi-user.target.wants/mariadb.service → /usr/lib                                                                    /systemd/system/mariadb.service.
[root@web1 ~]# ^C
[root@web1 ~]# mysql
MariaDB [(none)]> CREATE USER bi@'%' identified by '123';
Query OK, 0 rows affected (0.001 sec)

MariaDB [(none)]> GRANT ALL ON *.* TO bi@'%';
Query OK, 0 rows affected (0.001 sec)

MariaDB [(none)]> exit
Bye
[root@web1 ~]# mysql -ubi -p123
MariaDB [(none)]>

[root@web2 ~]# dnf install mariadb-server -y^
[root@web2 ~]# systemctl enable --now mariadb
[root@web2 ~]# systemctl enable --now mariadb
Created symlink /etc/systemd/system/mysql.service → /usr/lib/systemd/system/mariadb.se                                                                      rvice.
Created symlink /etc/systemd/system/mysqld.service → /usr/lib/systemd/system/mariadb.s                                                                      ervice.
Created symlink /etc/systemd/system/multi-user.target.wants/mariadb.service → /usr/lib                                                                      /systemd/system/mariadb.service.
[root@web2 ~]# mysql
MariaDB [(none)]> CREATE USER bi@'%' identified by '123';
Query OK, 0 rows affected (0.000 sec)

MariaDB [(none)]> GRANT ALL ON *.* TO bi@'%';
Query OK, 0 rows affected (0.001 sec)

MariaDB [(none)]> exit
Bye

[root@web2 ~]# mysql -ubi -p123

MariaDB [(none)]>

配置KA1 和 KA2 的第二个主备服务

bash 复制代码 
[root@KA1 ~]# vim /etc/keepalived/keepalived.conf
 32 vrrp_instance DB_VIP {
 33     state BACKUP
 34     interface eth0
 35     virtual_router_id 52
 36     priority 80
 37     davert_int 1
 38     authentication {
 39         auth_type PASS
 40         auth_pass 1111
 41     }
 42     virtual_ipaddress {
 43         192.168.248.200/24 dev eth0 label eth0:1
 44     }
 45 }

 77 virtual_server 192.168.248.200 3306 {
 78     delay_loop 6
 79     lb_algo rr
 80     lb_kind DR
 81     procotol TCP
 82
 83     real_server 192.168.248.10 3306 {
 84         weight 1
 85         HTTP_GET {
 86             url {
 87                 path /
 88                 status_code 200
 89                 }
 90             connect_timeout 1
 91             retry 3
 92             delay_before_retry 1
 93             }
 94     }
 95     real_server 192.168.248.20 3306 {
 96         weight  1
 97         TCP_CKECK {
 98             connect_timeout 5
 99             retry 3
100             delay_before_retry 3
101             connect_port 3306
102             }
103         }
104 }


[root@KA1 ~]# systemctl restart keepalived.service

[root@KA1 ~]# ip add
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    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
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether 00:0c:29:6d:7f:2b brd ff:ff:ff:ff:ff:ff
    altname enp3s0
    altname ens160
    inet 192.168.248.50/24 brd 192.168.248.255 scope global noprefixroute eth0
       valid_lft forever preferred_lft forever
    inet 192.168.248.100/24 scope global secondary eth0:0
       valid_lft forever preferred_lft forever
    inet6 fe80::7831:d38e:18e8:c266/64 scope link noprefixroute
       valid_lft forever preferred_lft forever
bash 复制代码 
[root@KA2 [root@KA2 ~]# systemctl restart keepalived.service
 34 vrrp_instance DB_VIP {
 35     state MASTER
 36     interface eth0
 37     virtual_router_id 52
 38     priority 100
 39     advert_int 1
 40     authentication {
 41         auth_type PASS
 42         auth_pass 1111
 43     }
 44     virtual_ipaddress {
 45         192.168.248.200/24 dev eth0 label eth0:1
 46     }
 47 }

 80 virtual_server 192.168.248.200 3306 {
 81     delay_loop 6
 82     lb_algo rr
 83     lb_kind DR
 84     protocol TCP
 85
 86     real_server 192.168.248.10 3306 {
 87         weight 1
 88         TCP_CHECK {
 89           connect_timeout 5
 90           retry 3
 91           delay_before_retry 3
 92           connect_port 3306
 93       }
 94     }
 95
 96     real_server 192.168.248.20 3306 {
 97         weight 1
 98         TCP_CHECK {
 99           connect_timeout 5
100           retry 3
101           delay_before_retry 3
102           connect_port 3306
103       }
104     }
105 }

[root@KA2 ~]# systemctl restart keepalived.service 

[root@KA2 ~]# ip add
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    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
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether 00:0c:29:f6:15:27 brd ff:ff:ff:ff:ff:ff
    altname enp3s0
    altname ens160
    inet 192.168.248.60/24 brd 192.168.248.255 scope global noprefixroute eth0
       valid_lft forever preferred_lft forever
    inet 192.168.248.200/24 scope global secondary eth0:1
       valid_lft forever preferred_lft forever
    inet6 fe80::3b5a:7484:2377:ed88/64 scope link noprefixroute
       valid_lft forever preferred_lft forever

3.3 测试

查看vip 挂载情况

bash 复制代码 
[root@KA1 ~]# ip add
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 10
    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
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen
    link/ether 00:0c:29:6d:7f:2b brd ff:ff:ff:ff:ff:ff
    altname enp3s0
    altname ens160
    inet 192.168.248.50/24 brd 192.168.248.255 scope global noprefixroute eth0
       valid_lft forever preferred_lft forever
    inet 192.168.248.100/24 scope global secondary eth0:0
       valid_lft forever preferred_lft forever
    inet6 fe80::7831:d38e:18e8:c266/64 scope link noprefixroute
       valid_lft forever preferred_lft forever

[root@KA2 ~]# ip add
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 10
    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
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen
    link/ether 00:0c:29:f6:15:27 brd ff:ff:ff:ff:ff:ff
    altname enp3s0
    altname ens160
    inet 192.168.248.60/24 brd 192.168.248.255 scope global noprefixroute eth0
       valid_lft forever preferred_lft forever
    inet 192.168.248.200/24 scope global secondary eth0:1
       valid_lft forever preferred_lft forever
    inet6 fe80::3b5a:7484:2377:ed88/64 scope link noprefixroute
       valid_lft forever preferred_lft forever

测试httpd 服务均衡负载情况

测试mysql 服务负载均衡情况（我在主机上登录了四次192.168.248.100的mysql服务）

关闭web1 mysql 服务 再链接 192.168.248.100 mysql服务三次

相关推荐
源远流长jerry3 小时前
DPDK MP (Multi-Process) 通道深度解析
linux·网络·架构·ip
RisunJan3 小时前
Linux命令-md5sum（计算和校验文件报文摘要的工具程序）
linux·运维
抹茶咖啡3 小时前
IT运维的365天--042 骚操作之--用IPSec给远程桌面上把锁
运维·网络·it运维
王琦03183 小时前
第三章 linux文件类型和根目录结构
linux·运维·服务器
minji...3 小时前
Linux 文件系统 (三) 软连接和硬链接
linux·运维·服务器·c++·算法
源远流长jerry4 小时前
DPDK 内存管理深度解析：从大页到 Mbuf 的完整链路
linux·服务器·网络
好像不对劲4 小时前
【docker】win10 wsl docker不能挂GPU
运维·docker·容器·wsl
leaves falling4 小时前
有效的字母异位词
java·服务器·前端
落日漫游5 小时前
MySQL双活与备份：高可用实战指南
运维
热门推荐
01GitHub 镜像站点02Qwen3.5 开源全解析：从 0.8B 到 397B，代际升级 + 全场景选型指南03小黑课堂计算机二级WPSoffice题库软件下载安装教程（2026年3月最新版）04围棋-html版本05班级宠物园部署指南06OpenClaw 使用和管理 MCP 完全指南07UV安装并设置国内源08OpenClaw Control UI安全上下文访问配置09“wsl --install -d Ubuntu-22.04”下载慢，中国地区离线安装 Ubuntu 22.04 WSL方法（亲测2025年5月6日）10Labelme从安装到标注：零基础完整指南