0
0

linux 网卡配置 vlan/bond/bridge/macvlan/ipvlan/macvtap 模式

linux 网卡模式

linux网卡支持非vlan模式、vlan模式、bond模式、bridge模式，macvlan模式、ipvlan模式等，下面介绍交换机端及服务器端配置示例。

前置要求：

  • 准备一台物理交换机，以 H3C S5130 三层交换机为例
  • 准备一台物理服务器，以 Ubuntu 22.04 LTS 操作系统为例

交换机创建2个示例VLAN，vlan10和vlan20，及VLAN接口。

  1. <H3C>system-view
  3. [H3C]vlan 1020[H3C]interface Vlan-interface 10[H3C-Vlan-interface10]ip address 172.16.10.1 24[H3C-Vlan-interface10]undo shutdown[H3C-Vlan-interface10]exit
  4. [H3C][H3C]interface Vlan-interface 20[H3C-Vlan-interface20]ip address 172.16.20.1 24[H3C-Vlan-interface20]undo  shutdown[H3C-Vlan-interface20]exit
  5. [H3C]

网卡非vlan模式

网卡非vlan模式，一般直接配置IP地址，对端上连交换机配置为access口，access口一般用于连接纯物理服务器或办公终端设备。

示意图如下
在这里插入图片描述

交换机配置，交换机接口配置为access模式，并加入对应vlan

  1. <H3C>system-view
  2. [H3C]interface GigabitEthernet 1/0/1
  3. [H3C-GigabitEthernet1/0/1]port link-type access
  4. [H3C-GigabitEthernet1/0/1]port access vlan 10[H3C-GigabitEthernet1/0/1]exit
  5. [H3C][H3C]interface GigabitEthernet 1/0/2
  6. [H3C-GigabitEthernet1/0/2]port link-type access
  7. [H3C-GigabitEthernet1/0/2]port access vlan 20[H3C-GigabitEthernet1/0/2]exit
  8. [H3C]

服务器1配置，服务器网卡直接配置IP地址

  1. root@server1:~# cat /etc/netplan/00-installer-config.yamlnetwork:ethernets:enp1s0:dhcp4:falseaddresses:- 172.16.10.10/24
  2.       nameservers:addresses:- 223.5.5.5
  3.           - 223.6.6.6
  4.       routes:-to: default
  5.           via: 172.16.10.1
  6.   version:2

服务器2配置，服务器网卡直接配置IP地址

  1. root@server2:~# cat /etc/netplan/00-installer-config.yamlnetwork:ethernets:enp1s0:dhcp4:falseaddresses:- 172.16.20.10/24
  2.       nameservers:addresses:- 223.5.5.5
  3.           - 223.6.6.6
  4.       routes:-to: default
  5.           via: 172.16.20.1
  6.   version:2

应用网卡配置

  1. netplan apply

查看服务器接口信息

  1. root@server1:~# ip a1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
  2.     link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
  3.     inet 127.0.0.1/8 scope host lo
  4.        valid_lft forever preferred_lft forever
  5.     inet6 ::1/128 scope host 
  6.        valid_lft forever preferred_lft forever
  7. 2: enp1s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
  8.     link/ether 7c:b5:9b:59:0a:71 brd ff:ff:ff:ff:ff:ff
  9.     inet 172.16.10.10/24 brd 172.16.10.255 scope global enp1s0
  10.        valid_lft forever preferred_lft forever
  11.     inet6 fe80::7eb5:9bff:fe59:a71/64 scope link 
  12.        valid_lft forever preferred_lft forever

通过server1 ping server2测试连通性，三层交换机支持路由功能，能够打通二层隔离的vlan网段。

  1. root@server1:~# ping 172.16.20.10 -c 4
  2. PING 172.16.20.10 (172.16.20.10)56(84) bytes of data.
  3. 64 bytes from 172.16.20.10: icmp_seq=1ttl=64time=0.033 ms
  4. 64 bytes from 172.16.20.10: icmp_seq=2ttl=64time=0.048 ms
  5. 64 bytes from 172.16.20.10: icmp_seq=3ttl=64time=0.048 ms
  6. 64 bytes from 172.16.20.10: icmp_seq=4ttl=64time=0.047 ms
  8. --- 172.16.20.10 ping statistics ---
  9. 4 packets transmitted, 4 received, 0% packet loss, time 3061ms
  10. rtt min/avg/max/mdev =0.033/0.044/0.048/0.006 ms

网卡vlan模式

vlan模式下，对端上连交换机需要配置为trunk口，允许多个vlan通过。

示意图如下
在这里插入图片描述

交换机配置，交换机需要配置为trunk口，允许多个vlan通过

  1. H3C>system-view 
  2. [H3C]interface GigabitEthernet 1/0/1
  3. [H3C-GigabitEthernet1/0/1]port link-type trunk
  4. [H3C-GigabitEthernet1/0/1]port trunk permit vlan 1020[H3C-GigabitEthernet1/0/1]exit
  5. [H3C]

服务器配置，服务器需要配置vlan子接口

  1. root@server1:~# cat /etc/netplan/00-installer-config.yamlnetwork:ethernets:enp1s0:dhcp4:truevlans:vlan10:id:10link: enp1s0
  2.       addresses:["172.16.10.10/24"]routes:-to: default
  3.           via: 172.16.10.1
  4.           metric:200vlan20:id:20link: enp1s0
  5.       addresses:["172.16.20.10/24"]routes:-to: default
  6.           via: 172.16.20.1
  7.           metric:300version:2

查看接口信息，新建了两个vlan子接口vlan10和vlan20

  1. root@server1:~# ip a1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
  2.     link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
  3.     inet 127.0.0.1/8 scope host lo
  4.        valid_lft forever preferred_lft forever
  5.     inet6 ::1/128 scope host 
  6.        valid_lft forever preferred_lft forever
  7. 2: enp1s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
  8.     link/ether 7c:b5:9b:59:0a:71 brd ff:ff:ff:ff:ff:ff
  9.     inet6 fe80::7eb5:9bff:fe59:a71/64 scope link 
  10.        valid_lft forever preferred_lft forever
  11. 10: vlan10@enp1s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
  12.     link/ether 7c:b5:9b:59:0a:71 brd ff:ff:ff:ff:ff:ff
  13.     inet 172.16.10.10/24 brd 172.16.10.255 scope global vlan10
  14.        valid_lft forever preferred_lft forever
  15.     inet6 fe80::7eb5:9bff:fe59:a71/64 scope link 
  16.        valid_lft forever preferred_lft forever
  17. 11: vlan20@enp1s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
  18.     link/ether 7c:b5:9b:59:0a:71 brd ff:ff:ff:ff:ff:ff
  19.     inet 172.16.20.10/24 brd 172.16.20.255 scope global vlan20
  20.        valid_lft forever preferred_lft forever
  21.     inet6 fe80::7eb5:9bff:fe59:a71/64 scope link 
  22.        valid_lft forever preferred_lft forever

通过vlan10 和 vlan20测试与网关连通性

  1. root@server1:~# ping 172.16.10.1 -c 4
  2. PING 172.16.10.1 (172.16.10.1)56(84) bytes of data.
  3. 64 bytes from 172.16.10.1: icmp_seq=1ttl=64time=0.033 ms
  4. 64 bytes from 172.16.10.1: icmp_seq=2ttl=64time=0.048 ms
  5. 64 bytes from 172.16.10.1: icmp_seq=3ttl=64time=0.048 ms
  6. 64 bytes from 172.16.10.1: icmp_seq=4ttl=64time=0.047 ms
  8. --- 172.16.10.1 ping statistics ---
  9. 4 packets transmitted, 4 received, 0% packet loss, time 3061ms
  10. rtt min/avg/max/mdev =0.033/0.044/0.048/0.006 ms
  11. root@server1:~#
  12. root@server1:~# ping 172.16.20.1 -c 4
  13. PING 172.16.20.1 (172.16.20.1)56(84) bytes of data.
  14. 64 bytes from 172.16.20.1: icmp_seq=1ttl=64time=0.033 ms
  15. 64 bytes from 172.16.20.1: icmp_seq=2ttl=64time=0.048 ms
  16. 64 bytes from 172.16.20.1: icmp_seq=3ttl=64time=0.048 ms
  17. 64 bytes from 172.16.20.1: icmp_seq=4ttl=64time=0.047 ms
  19. --- 172.16.20.1 ping statistics ---
  20. 4 packets transmitted, 4 received, 0% packet loss, time 3061ms
  21. rtt min/avg/max/mdev =0.033/0.044/0.048/0.006 ms

网卡bond模式

bond模式下，对端交换机需要配置bond聚合口。

示意图如下
在这里插入图片描述

交换机配置，配置动态链路聚合，将端口g1/0/1和g1/0/3加入聚合组。然后将bond口配置为trunk模式。

  1. <H3C>system-view
  2. [H3C]interface Bridge-Aggregation 1[H3C-Bridge-Aggregation1]link-aggregation mode dynamic
  3. [H3C-Bridge-Aggregation1]quit
  5. [H3C]interface GigabitEthernet 1/0/1
  6. [H3C-GigabitEthernet1/0/1]port link-aggregation group 1[H3C-GigabitEthernet1/0/1]exit
  8. [H3C]interface GigabitEthernet 1/0/3
  9. [H3C-GigabitEthernet1/0/3]port link-aggregation group 1[H3C-GigabitEthernet1/0/3]exit
  11. [H3C]interface Bridge-Aggregation 1[H3C-Bridge-Aggregation1]port link-type trunk
  12. [H3C-Bridge-Aggregation1]port trunk permit vlan 1020[H3C-Bridge-Aggregation1]exit

服务器配置

  1. root@server1:~# cat /etc/netplan/00-installer-config.yamlnetwork:version:2ethernets:enp1s0:dhcp4: no
  2.     enp2s0:dhcp4: no
  3.   bonds:bond0:interfaces:- enp1s0
  4.         - enp2s0
  5.       parameters:mode: 802.3ad
  6.         lacp-rate: fast
  7.         mii-monitor-interval:100transmit-hash-policy: layer2+3
  8.   vlans:vlan10:id:10link: bond0
  9.       addresses:["172.16.10.10/24"]routes:-to: default
  10.           via: 172.16.10.1
  11.           metric:200vlan20:id:20link: bond0
  12.       addresses:["172.16.20.10/24"]routes:-to: default
  13.           via: 172.16.20.1
  14.           metric:300

查看网卡信息，新建了bond0网口，并且基于bond0网口创建了两个vlan子接口vlan10和vlan20，enp1s0和enp2s0显示

  1. master bond0

，说明两个网卡属于bond0成员接口。

  1. root@server1:~# ip a1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
  2.     link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
  3.     inet 127.0.0.1/8 scope host lo
  4.        valid_lft forever preferred_lft forever
  5.     inet6 ::1/128 scope host 
  6.        valid_lft forever preferred_lft forever
  7. 2: enp1s0: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 1500 qdisc fq_codel master bond0 state UP group default qlen 1000
  8.     link/ether ae:fd:60:48:84:1a brd ff:ff:ff:ff:ff:ff permaddr 7c:b5:9b:59:0a:71
  9. 3: enp2s0: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 1500 qdisc fq_codel master bond0 state UP group default qlen 1000
  10.     link/ether ae:fd:60:48:84:1a brd ff:ff:ff:ff:ff:ff permaddr e4:54:e8:dc:e5:88
  11. 7: bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
  12.     link/ether ae:fd:60:48:84:1a brd ff:ff:ff:ff:ff:ff
  13.     inet6 fe80::acfd:60ff:fe48:841a/64 scope link 
  14.        valid_lft forever preferred_lft forever
  15. 8: vlan10@bond0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
  16.     link/ether ae:fd:60:48:84:1a brd ff:ff:ff:ff:ff:ff
  17.     inet 172.16.10.10/24 brd 172.16.10.255 scope global vlan10
  18.        valid_lft forever preferred_lft forever
  19.     inet6 fe80::acfd:60ff:fe48:841a/64 scope link 
  20.        valid_lft forever preferred_lft forever
  21. 9: vlan20@bond0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
  22.     link/ether ae:fd:60:48:84:1a brd ff:ff:ff:ff:ff:ff
  23.     inet 172.16.20.10/24 brd 172.16.20.255 scope global vlan20
  24.        valid_lft forever preferred_lft forever
  25.     inet6 fe80::acfd:60ff:fe48:841a/64 scope link 
  26.        valid_lft forever preferred_lft forever

查看bond状态，

  1. Bonding Mode

显示为

  1. IEEE 802.3ad Dynamic link aggregation

，并且下面

  1. Slave Interface

显示了两个成员接口的信息。

  1. root@server1:~# cat /proc/net/bonding/bond0Ethernet Channel Bonding Driver: v5.15.0-60-generic
  3. Bonding Mode: IEEE 802.3ad Dynamic link aggregation
  4. Transmit Hash Policy: layer2+3 (2)
  5. MII Status: up
  6. MII Polling Interval (ms):100Up Delay (ms):0Down Delay (ms):0Peer Notification Delay (ms):0
  8. 802.3ad info
  9. LACP active: on
  10. LACP rate: fast
  11. Min links:0Aggregator selection policy (ad_select): stable
  12. System priority:65535System MAC address: ae:fd:60:48:84:1a
  13. Active Aggregator Info:Aggregator ID:1Number of ports:2Actor Key:9Partner Key:1Partner Mac Address: fc:60:9b:35:ad:18Slave Interface: enp1s0
  14. MII Status: up
  15. Speed: 1000 Mbps
  16. Duplex: full
  17. Link Failure Count:2Permanent HW addr: 7c:b5:9b:59:0a:71Slave queue ID:0Aggregator ID:1Actor Churn State: none
  18. Partner Churn State: none
  19. Actor Churned Count:0Partner Churned Count:0details actor lacp pdu:system priority:65535system mac address: ae:fd:60:48:84:1a
  20.     port key:9port priority:255port number:1port state:63details partner lacp pdu:system priority:32768system mac address: fc:60:9b:35:ad:18oper key:1port priority:32768port number:2port state:61Slave Interface: enp2s0
  21. MII Status: up
  22. Speed: 1000 Mbps
  23. Duplex: full
  24. Link Failure Count:3Permanent HW addr: e4:54:e8:dc:e5:88Slave queue ID:0Aggregator ID:1Actor Churn State: none
  25. Partner Churn State: none
  26. Actor Churned Count:0Partner Churned Count:0details actor lacp pdu:system priority:65535system mac address: ae:fd:60:48:84:1a
  27.     port key:9port priority:255port number:2port state:63details partner lacp pdu:system priority:32768system mac address: fc:60:9b:35:ad:18oper key:1port priority:32768port number:1port state:61

测试连通性，测试与交换机网关地址的连通性：

  1. root@server1:~# ping 172.16.10.1 -c 4
  2. PING 172.16.10.1 (172.16.10.1)56(84) bytes of data.
  3. 64 bytes from 172.16.10.1: icmp_seq=1ttl=255time=1.64 ms
  4. 64 bytes from 172.16.10.1: icmp_seq=2ttl=255time=1.59 ms
  5. 64 bytes from 172.16.10.1: icmp_seq=3ttl=255time=1.95 ms
  6. 64 bytes from 172.16.10.1: icmp_seq=4ttl=255time=1.93 ms
  8. --- 172.16.10.1 ping statistics ---
  9. 4 packets transmitted, 4 received, 0% packet loss, time 3006ms
  10. rtt min/avg/max/mdev =1.589/1.776/1.953/0.165 ms
  11. root@server1:~#

关闭一个接口，再次测试连通性，依然能够ping通

  1. root@server1:~# ip link set dev enp2s0 down
  2. root@server1:~# ip link show enp2s03: enp2s0: <BROADCAST,MULTICAST,SLAVE> mtu 1500 qdisc fq_codel master bond0 state DOWN mode DEFAULT group default qlen 1000
  3.     link/ether ae:fd:60:48:84:1a brd ff:ff:ff:ff:ff:ff permaddr e4:54:e8:dc:e5:88
  4. root@server1:~# 
  5. root@server1:~# ping 172.16.10.1 -c 4
  6. PING 172.16.10.1 (172.16.10.1)56(84) bytes of data.
  7. 64 bytes from 172.16.10.1: icmp_seq=1ttl=255time=1.54 ms
  8. 64 bytes from 172.16.10.1: icmp_seq=2ttl=255time=1.64 ms
  9. 64 bytes from 172.16.10.1: icmp_seq=3ttl=255time=2.73 ms
  10. 64 bytes from 172.16.10.1: icmp_seq=4ttl=255time=1.47 ms
  12. --- 172.16.10.1 ping statistics ---
  13. 4 packets transmitted, 4 received, 0% packet loss, time 3006ms
  14. rtt min/avg/max/mdev =1.470/1.844/2.732/0.516 ms

网卡桥接模式

桥接模式下，对端交换机可配置access模式或trunk模式。

示意图如下

在这里插入图片描述

交换机配置，交换机接口配置为access模式为例，并加入对应vlan

  1. <H3C>system-view
  2. [H3C]interface GigabitEthernet 1/0/1
  3. [H3C-GigabitEthernet1/0/1]port link-type access
  4. [H3C-GigabitEthernet1/0/1]port access vlan 10[H3C-GigabitEthernet1/0/1]exit
  5. [H3C]

服务器配置，物理网卡加入到网桥中，IP地址配置到网桥接口br0上。

  1. root@server1:~# cat /etc/netplan/00-installer-config.yamlnetwork:version:2ethernets:enp1s0:dhcp4: no
  2.       dhcp6: no
  3.   bridges:br0:interfaces:[enp1s0]addresses:[172.16.10.10/24]routes:-to: default
  4.         via: 172.16.10.1
  5.         metric:100on-link:truemtu:1500nameservers:addresses:- 223.5.5.5
  6.           - 223.6.6.6
  7.       parameters:stp:trueforward-delay:4

查看网卡信息

  1. root@server1:~# ip a1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
  2.     link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
  3.     inet 127.0.0.1/8 scope host lo
  4.        valid_lft forever preferred_lft forever
  5.     inet6 ::1/128 scope host 
  6.        valid_lft forever preferred_lft forever
  7. 2: enp1s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel master br0 state UP group default qlen 1000
  8.     link/ether 7c:b5:9b:59:0a:71 brd ff:ff:ff:ff:ff:ff
  9. 12: br0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
  10.     link/ether 0e:d0:7e:31:9c:74 brd ff:ff:ff:ff:ff:ff
  11.     inet 172.16.10.10/24 brd 172.16.10.255 scope global br0
  12.        valid_lft forever preferred_lft forever
  13.     inet6 fe80::cd0:7eff:fe31:9c74/64 scope link 
  14.        valid_lft forever preferred_lft forever

查看网桥及接口，当前网桥上只有一个物理接口enp1s0。

  1. root@server1:~# apt install -y bridge-utils
  2. root@ubuntu:~# brctl show
  3. bridge name     bridge id               STP enabled     interfaces
  4. br0             8000.0ed07e319c74       yes             enp1s0
  5. root@server1:~#

这样在KVM虚拟化环境，虚拟机实例连接到网桥后，虚拟机可以配置与物理网卡相同网段的IP地址。访问虚拟机可以像访问物理机一样方便。

网卡macvlan模式

macvlan(MAC Virtual LAN)是Linux内核提供的一种网络虚拟化技术，它允许在一个物理网卡接口上创建多个虚拟网卡接口，每个虚拟接口都有自己独立的MAC地址，也可以配置上 IP 地址进行通信。Macvlan 下的虚拟机或者容器网络和主机在同一个网段中，共享同一个广播域。

macvlan模式下，对端交换机可配置access模式或trunk模式，trunk模式下macvlan能够与vlan很好的结合使用。

示意图如下：

在这里插入图片描述

macvlan IP模式

该模式下，上连交换机接口配置为access模式，服务器macvlan主网卡和子接口直接配置相同网段的IP地址。

交换机配置

  1. <H3C>system-view
  2. [H3C]interface GigabitEthernet 1/0/1
  3. [H3C-GigabitEthernet1/0/1]port link-type access
  4. [H3C-GigabitEthernet1/0/1]port access vlan 10[H3C-GigabitEthernet1/0/1]exit

服务器配置，macvlan支持多种模式，这里使用bridge模式，并持久化配置

  1. cat>/etc/networkd-dispatcher/routable.d/10-macvlan-interfaces.sh<<EOF
  2. #! /bin/bash
  3. ip link add macvlan0 link enp1s0 type macvlan mode bridge
  4. ip link add macvlan1 link enp1s0 type macvlan mode bridge
  5. EOFchmod o+x,g+x,u+x /etc/networkd-dispatcher/routable.d/10-macvlan-interfaces.sh

配置netplan

  1. root@server1:~# cat /etc/netplan/00-installer-config.yamlnetwork:ethernets:enp1s0:dhcp4:falseaddresses:- 172.16.10.10/24
  2.       nameservers:addresses:- 223.5.5.5
  3.           - 223.6.6.6
  4.       routes:-to: default
  5.           via: 172.16.10.1
  6.     macvlan0:addresses:- 172.16.10.11/24
  7.     macvlan1:addresses:- 172.16.10.12/24
  8.   version:2

应用网卡配置

  1. netplan apply

查看网卡信息，新建了两个macvlan接口，IP地址与主网卡位于同一网段，并且每个接口都有独立的MAC地址。

  1. root@server1:~# ip a1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
  2.     link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
  3.     inet 127.0.0.1/8 scope host lo
  4.        valid_lft forever preferred_lft forever
  5.     inet6 ::1/128 scope host 
  6.        valid_lft forever preferred_lft forever
  7. 2: enp1s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
  8.     link/ether 7c:b5:9b:59:0a:71 brd ff:ff:ff:ff:ff:ff
  9.     inet 172.16.10.10/24 brd 172.16.10.255 scope global enp1s0
  10.        valid_lft forever preferred_lft forever
  11. 13: macvlan0@enp1s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
  12.     link/ether 32:e8:b4:0a:47:62 brd ff:ff:ff:ff:ff:ff
  13.     inet 172.16.10.11/24 brd 172.16.10.255 scope global macvlan0
  14.        valid_lft forever preferred_lft forever
  15.     inet6 fe80::30e8:b4ff:fe0a:4762/64 scope link 
  16.        valid_lft forever preferred_lft forever
  17. 14: macvlan1@enp1s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
  18.     link/ether d2:73:75:14:b2:04 brd ff:ff:ff:ff:ff:ff
  19.     inet 172.16.10.12/24 brd 172.16.10.255 scope global macvlan1
  20.        valid_lft forever preferred_lft forever
  21.     inet6 fe80::d073:75ff:fe14:b204/64 scope link 
  22.        valid_lft forever preferred_lft forever

测试与网关的连通性

  1. root@server1:~# ping -c 3 172.16.10.1
  2. PING 172.16.10.1 (172.16.10.1)56(84) bytes of data.
  3. 64 bytes from 172.16.10.1: icmp_seq=1ttl=255time=3.60 ms
  4. 64 bytes from 172.16.10.1: icmp_seq=2ttl=255time=1.45 ms
  5. 64 bytes from 172.16.10.1: icmp_seq=3ttl=255time=1.44 ms
  7. --- 172.16.10.1 ping statistics ---
  8. 3 packets transmitted, 3 received, 0% packet loss, time 2004ms
  9. rtt min/avg/max/mdev =1.441/2.163/3.602/1.017 ms
  10. root@server1:~#

macvlan vlan模式

该模式下，上连交换机接口配置为trunk模式，服务器macvlan主网卡不配置IP地址，每个macvlan子接口配置为不同的vlan子接口。

交换机配置

  1. <H3C>system-view
  2. [H3C]interface GigabitEthernet 1/0/1
  3. [H3C-GigabitEthernet1/0/3]port link-type trunk
  4. [H3C-GigabitEthernet1/0/3]port trunk permit vlan 1020[H3C-GigabitEthernet1/0/1]exit
  5. [H3C]

服务器配置，macvlan支持多种模式，这里使用bridge模式，并持久化配置

  1. cat>/etc/networkd-dispatcher/routable.d/10-macvlan-interfaces.sh<<EOF
  2. #! /bin/bash
  3. ip link add macvlan0 link enp1s0 type macvlan mode bridge
  4. ip link add macvlan1 link enp1s0 type macvlan mode bridge
  5. EOFchmod o+x,g+x,u+x /etc/networkd-dispatcher/routable.d/10-macvlan-interfaces.sh

配置netplan，两个macvlan接口macvlan0和macvlan1分别配置vlan子接口vlan10和vlan20。

  1. root@ubuntu:~# cat /etc/netplan/00-installer-config.yamlnetwork:ethernets:enp1s0:dhcp4:falsemacvlan0:dhcp4:falsemacvlan1:dhcp4:falsevlans:vlan10:id:10link: macvlan0
  2.       addresses:["172.16.10.10/24"]routes:-to: default
  3.           via: 172.16.10.1
  4.           metric:200vlan20:id:20link: macvlan1
  5.       addresses:["172.16.20.10/24"]routes:-to: default
  6.           via: 172.16.20.1
  7.           metric:300version:2

应用网卡配置

  1. netplan apply

查看网卡信息，新建了两个macvlan接口，以及对应的两个vlan子接口。

  1. root@server1:~# ip a1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
  2.     link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
  3.     inet 127.0.0.1/8 scope host lo
  4.        valid_lft forever preferred_lft forever
  5.     inet6 ::1/128 scope host 
  6.        valid_lft forever preferred_lft forever
  7. 2: enp1s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
  8.     link/ether 7c:b5:9b:59:0a:71 brd ff:ff:ff:ff:ff:ff
  9.     inet6 fe80::7eb5:9bff:fe59:a71/64 scope link 
  10.        valid_lft forever preferred_lft forever
  11. 11: macvlan0@enp1s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
  12.     link/ether 32:e8:b4:0a:47:62 brd ff:ff:ff:ff:ff:ff
  13.     inet6 fe80::30e8:b4ff:fe0a:4762/64 scope link 
  14.        valid_lft forever preferred_lft forever
  15. 12: macvlan1@enp1s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
  16.     link/ether d2:73:75:14:b2:04 brd ff:ff:ff:ff:ff:ff
  17.     inet6 fe80::d073:75ff:fe14:b204/64 scope link 
  18.        valid_lft forever preferred_lft forever
  19. 13: vlan10@macvlan0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
  20.     link/ether 32:e8:b4:0a:47:62 brd ff:ff:ff:ff:ff:ff
  21.     inet 172.16.10.10/24 brd 172.16.10.255 scope global vlan10
  22.        valid_lft forever preferred_lft forever
  23.     inet6 fe80::30e8:b4ff:fe0a:4762/64 scope link 
  24.        valid_lft forever preferred_lft forever
  25. 14: vlan20@macvlan1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
  26.     link/ether d2:73:75:14:b2:04 brd ff:ff:ff:ff:ff:ff
  27.     inet 172.16.20.10/24 brd 172.16.20.255 scope global vlan20
  28.        valid_lft forever preferred_lft forever
  29.     inet6 fe80::d073:75ff:fe14:b204/64 scope link 
  30.        valid_lft forever preferred_lft forever

测试两个VLAN接口与外部网关的连通性

  1. root@server1:~# ping -c 3 172.16.10.1
  2. PING 172.16.10.1 (172.16.10.1)56(84) bytes of data.
  3. 64 bytes from 172.16.10.1: icmp_seq=1ttl=255time=3.60 ms
  4. 64 bytes from 172.16.10.1: icmp_seq=2ttl=255time=1.45 ms
  5. 64 bytes from 172.16.10.1: icmp_seq=3ttl=255time=1.44 ms
  7. --- 172.16.10.1 ping statistics ---
  8. 3 packets transmitted, 3 received, 0% packet loss, time 2004ms
  9. rtt min/avg/max/mdev =1.441/2.163/3.602/1.017 ms
  10. root@server1:~# 
  11. root@server1:~# ping -c 3 172.16.20.1 
  12. PING 172.16.20.1 (172.16.20.1)56(84) bytes of data.
  13. 64 bytes from 172.16.20.1: icmp_seq=1ttl=255time=1.35 ms
  14. 64 bytes from 172.16.20.1: icmp_seq=2ttl=255time=1.48 ms
  15. 64 bytes from 172.16.20.1: icmp_seq=3ttl=255time=1.46 ms
  17. --- 172.16.20.1 ping statistics ---
  18. 3 packets transmitted, 3 received, 0% packet loss, time 2004ms
  19. rtt min/avg/max/mdev =1.353/1.429/1.477/0.054 ms
  20. root@server1:~#

网卡ipvlan模式

IPVLAN(IP Virtual LAN)是Linux内核提供的一种网络虚拟化技术，它可以在一个物理网卡上创建多个虚拟网卡接口，每个虚拟接口都有自己独立的IP地址。

IPVLAN和macvlan类似，都是从一个主机接口虚拟出多个虚拟网络接口。唯一比较大的区别就是ipvlan虚拟出的子接口都有相同的mac地址（与物理接口共用同个mac地址），但可配置不同的ip地址。

ipvlan模式下，对端交换机也可以配置access模式或trunk模式，trunk模式下ipvlan能够与vlan很好的结合使用。

示意图如下
在这里插入图片描述

交换机配置

  1. <H3C>system-view
  2. [H3C]interface GigabitEthernet 1/0/1
  3. [H3C-GigabitEthernet1/0/1]port link-type access
  4. [H3C-GigabitEthernet1/0/1]port access vlan 10[H3C-GigabitEthernet1/0/1]exit
  5. [H3C]

服务器配置，ipvlan支持三种模式(l2、l3、l3s)，这里使用l3模式，并持久化配置

  1. cat>/etc/networkd-dispatcher/routable.d/10-ipvlan-interfaces.sh<<EOF
  2. #! /bin/bash
  3. ip link add ipvlan0 link enp1s0 type ipvlan mode l3
  4. ip link add ipvlan1 link enp1s0 type ipvlan mode l3
  5. EOFchmod o+x,g+x,u+x /etc/networkd-dispatcher/routable.d/10-ipvlan-interfaces.sh

配置netplan

  1. root@server1:~# cat /etc/netplan/00-installer-config.yamlnetwork:ethernets:enp1s0:dhcp4:falseaddresses:- 172.16.10.10/24
  2.       nameservers:addresses:- 223.5.5.5
  3.           - 223.6.6.6
  4.       routes:-to: default
  5.           via: 172.16.10.1
  6.     ipvlan0:addresses:- 172.16.10.11/24
  7.     ipvlan1:addresses:- 172.16.10.12/24
  8.   version:2

应用网卡配置

  1. netplan apply

查看网卡信息，新建了两ipvlan接口，IP地址与主网卡位于同一网段，并且每个接口都有与主网卡相同的MAC地址。

  1. root@server1:~# ip a1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
  2.     link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
  3.     inet 127.0.0.1/8 scope host lo
  4.        valid_lft forever preferred_lft forever
  5.     inet6 ::1/128 scope host 
  6.        valid_lft forever preferred_lft forever
  7. 2: enp1s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
  8.     link/ether 7c:b5:9b:59:0a:71 brd ff:ff:ff:ff:ff:ff
  9.     inet 172.16.10.10/24 brd 172.16.10.255 scope global enp1s0
  10.        valid_lft forever preferred_lft forever
  11.     inet6 fe80::7eb5:9bff:fe59:a71/64 scope link 
  12.        valid_lft forever preferred_lft forever
  13. 9: ipvlan0@enp1s0: <BROADCAST,MULTICAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN group default qlen 1000
  14.     link/ether 7c:b5:9b:59:0a:71 brd ff:ff:ff:ff:ff:ff
  15.     inet 172.16.10.11/24 brd 172.16.10.255 scope global ipvlan0
  16.        valid_lft forever preferred_lft forever
  17.     inet6 fe80::7cb5:9b00:159:a71/64 scope link 
  18.        valid_lft forever preferred_lft forever
  19. 10: ipvlan1@enp1s0: <BROADCAST,MULTICAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN group default qlen 1000
  20.     link/ether 7c:b5:9b:59:0a:71 brd ff:ff:ff:ff:ff:ff
  21.     inet 172.16.10.12/24 brd 172.16.10.255 scope global ipvlan1
  22.        valid_lft forever preferred_lft forever
  23.     inet6 fe80::7cb5:9b00:259:a71/64 scope link 
  24.        valid_lft forever preferred_lft forever

测试与网关的连通性

  1. root@server1:~# ping -c 3 172.16.10.1
  2. PING 172.16.10.1 (172.16.10.1)56(84) bytes of data.
  3. 64 bytes from 172.16.10.1: icmp_seq=1ttl=255time=3.60 ms
  4. 64 bytes from 172.16.10.1: icmp_seq=2ttl=255time=1.45 ms
  5. 64 bytes from 172.16.10.1: icmp_seq=3ttl=255time=1.44 ms
  7. --- 172.16.10.1 ping statistics ---
  8. 3 packets transmitted, 3 received, 0% packet loss, time 2004ms
  9. rtt min/avg/max/mdev =1.441/2.163/3.602/1.017 ms
  10. root@server1:~#

网卡 macvtap 模式

使用 bridge 使 KVM 虚拟机能够进行外部通信的另一种替代方法是使用 Linux MacVTap 驱动程序。当不想创建普通网桥，但希望本地网络中的用户访问虚拟机时，可以使用 MacVTap。

与使用bridge 的一个主要区别是 MacVTap 直接连接到 KVM 主机中的网络接口。这种直接连接绕过了 KVM 主机中与连接和使用软件bridge 相关的大部分代码和组件，有效地缩短了代码路径。这种较短的代码路径通常会提高吞吐量并减少外部系统的延迟。

示意图如下：
在这里插入图片描述

交换机配置

  1. <H3C>system-view
  2. [H3C]interface GigabitEthernet 1/0/1
  3. [H3C-GigabitEthernet1/0/1]port link-type access
  4. [H3C-GigabitEthernet1/0/1]port access vlan 10[H3C-GigabitEthernet1/0/1]exit

主机网卡配置

  1. root@server1:~# cat /etc/netplan/00-installer-config.yaml# This is the network config written by 'subiquity'network:ethernets:enp1s0:dhcp4:falseaddresses:- 172.16.10.10/24
  2.       nameservers:addresses:- 223.5.5.5
  3.           - 223.6.6.6
  4.       routes:-to: default
  5.           via: 192.168.137.2
  6.   version:2

安装kvm虚拟化环境，创建两个虚拟机，指定从enp1s0主网卡分配mavtap子接口。

  1. virt-install \--name vm1 \--vcpus1\--memory2048\--diskpath=/var/lib/libvirt/images/vm1/jammy-server-cloudimg-amd64.img \
  2.   --os-variant ubuntu22.04 \--noautoconsole\--import\--autostart\--networktype=direct,source=enp1s0,source_mode=bridge,model=virtio
  4. virt-install \--name vm2 \--vcpus1\--memory2048\--diskpath=/var/lib/libvirt/images/vm2/jammy-server-cloudimg-amd64.img \
  5.   --os-variant ubuntu22.04 \--noautoconsole\--import\--autostart\--networktype=direct,source=enp1s0,source_mode=bridge,model=virtio

查看网卡信息，新创建了两个macvtap接口

  1. root@server1:~# ip a1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
  2.     link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
  3.     inet 127.0.0.1/8 scope host lo
  4.        valid_lft forever preferred_lft forever
  5.     inet6 ::1/128 scope host 
  6.        valid_lft forever preferred_lft forever
  7. 2: enp1s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
  8.     link/ether 7c:b5:9b:59:0a:71 brd ff:ff:ff:ff:ff:ff
  9.     inet 172.16.10.10/24 brd 172.16.10.255 scope global enp1s0
  10.        valid_lft forever preferred_lft forever
  11.     inet6 fe80::7eb5:9bff:fe59:a71/64 scope link 
  12.        valid_lft forever preferred_lft forever
  13. 5: virbr0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN group default qlen 1000
  14.     link/ether 52:54:00:bb:15:22 brd ff:ff:ff:ff:ff:ff
  15.     inet 192.168.122.1/24 brd 192.168.122.255 scope global virbr0
  16.        valid_lft forever preferred_lft forever
  17. 6: macvtap0@enp1s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 500
  18.     link/ether 52:54:00:41:8f:a3 brd ff:ff:ff:ff:ff:ff
  19.     inet6 fe80::5054:ff:fe41:8fa3/64 scope link 
  20.        valid_lft forever preferred_lft forever
  21. 7: macvtap1@enp1s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 500
  22.     link/ether 52:54:00:93:2c:4a brd ff:ff:ff:ff:ff:ff
  23.     inet6 fe80::5054:ff:fe93:2c4a/64 scope link 
  24.        valid_lft forever preferred_lft forever

虚拟机1配置IP地址

  1. root@vm1:~# cat /etc/netplan/00-installer-config.yamlnetwork:ethernets:enp1s0:dhcp4:falseaddresses:- 172.16.10.11/24
  2.       nameservers:addresses:- 223.5.5.5
  3.           - 223.6.6.6
  4.       routes:-to: default
  5.           via: 172.16.10.1
  6.   version:2

虚拟机2配置IP地址

  1. root@vm2:~# cat /etc/netplan/00-installer-config.yamlnetwork:ethernets:enp1s0:dhcp4:falseaddresses:- 172.16.10.12/24
  2.       nameservers:addresses:- 223.5.5.5
  3.           - 223.6.6.6
  4.       routes:-to: default
  5.           via: 172.16.10.1
  6.   version:2

测试与网关的连通性

  1. root@vm1:~# ping 172.16.10.1 -c 3
  2. PING 172.16.10.1 (172.16.10.1)56(84) bytes of data.
  3. 64 bytes from 172.16.10.1: icmp_seq=1ttl=255time=1.38 ms
  4. 64 bytes from 172.16.10.1: icmp_seq=2ttl=255time=1.75 ms
  5. 64 bytes from 172.16.10.1: icmp_seq=3ttl=255time=4.34 ms
  7. --- 172.16.10.1 ping statistics ---
  8. 3 packets transmitted, 3 received, 0% packet loss, time 2004ms
  9. rtt min/avg/max/mdev =1.382/2.491/4.344/1.318 ms

bond、vlan与桥接混合配置

将服务器两块网卡组成bond口，在bond口之上创建两个vlan子接口，分别加入两个linux bridge中，然后在不同bridge下创建虚拟机，虚拟机将属于不同vlan。

示意图如下：
在这里插入图片描述
交换机配置，配置动态链路聚合，将端口g1/0/1和g1/0/3加入聚合组。将聚合口配置为trunk模式，允许vlan 8 10 20通过，并且将vlan8 配置为聚合口的native vlan，作为管理使用。

  1. <H3C>system-view
  2. [H3C]interface Vlan-interface 8[H3C-Vlan-interface8]ip address 172.16.8.1 24[H3C-Vlan-interface8]exit
  3. [H3C][H3C]interface Bridge-Aggregation 1[H3C-Bridge-Aggregation1]link-aggregation mode dynamic
  4. [H3C-Bridge-Aggregation1]quit
  6. [H3C]interface GigabitEthernet 1/0/1
  7. [H3C-GigabitEthernet1/0/1]port link-aggregation group 1[H3C-GigabitEthernet1/0/1]exit
  9. [H3C]interface GigabitEthernet 1/0/3
  10. [H3C-GigabitEthernet1/0/3]port link-aggregation group 1[H3C-GigabitEthernet1/0/3]exit
  12. [H3C]interface Bridge-Aggregation 1[H3C-Bridge-Aggregation1]port link-type trunk 
  13. [H3C-Bridge-Aggregation1]port trunk permit vlan 81020[H3C-Bridge-Aggregation1]port trunk pvid vlan 8[H3C-Bridge-Aggregation1]undo port trunk permit vlan 1[H3C-Bridge-Aggregation1]exit
  14. [H3C]

服务器网卡配置，注意bond0配置了管理IP地址，匹配交换机native vlan 8。

  1. root@server1:~# cat /etc/netplan/00-installer-config.yamlnetwork:version:2ethernets:enp1s0:dhcp4:falseenp2s0:dhcp4:falsebonds:bond0:dhcp4:falsedhcp6:falseinterfaces:- enp1s0
  2.         - enp2s0
  3.       addresses:- 172.16.8.10/24
  4.       nameservers:addresses:- 223.5.5.5
  5.           - 223.6.6.6
  6.       routes:-to: default
  7.           via: 172.16.8.1
  8.       parameters:mode: 802.3ad
  9.         lacp-rate: fast
  10.         mii-monitor-interval:100transmit-hash-policy: layer2+3
  11.   bridges:br10:interfaces:[ vlan10 ]br20:interfaces:[ vlan20 ]vlans:vlan10:id:10link: bond0
  12.     vlan20:id:20link: bond0

查看网卡信息，新建了bond0网口，并且基于bond0网口创建了两个vlan子接口vlan10和vlan20。

  1. root@server1:~# ip a1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
  2.     link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
  3.     inet 127.0.0.1/8 scope host lo
  4.        valid_lft forever preferred_lft forever
  5.     inet6 ::1/128 scope host 
  6.        valid_lft forever preferred_lft forever
  7. 2: enp1s0: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 1500 qdisc fq_codel master bond0 state UP group default qlen 1000
  8.     link/ether ae:fd:60:48:84:1a brd ff:ff:ff:ff:ff:ff permaddr 7c:b5:9b:59:0a:71
  9. 3: enp2s0: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 1500 qdisc fq_codel master bond0 state UP group default qlen 1000
  10.     link/ether ae:fd:60:48:84:1a brd ff:ff:ff:ff:ff:ff permaddr e4:54:e8:dc:e5:88
  11. 15: bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
  12.     link/ether ae:fd:60:48:84:1a brd ff:ff:ff:ff:ff:ff
  13.     inet 172.16.8.10/24 brd 172.16.8.255 scope global bond0
  14.        valid_lft forever preferred_lft forever
  15.     inet6 fe80::acfd:60ff:fe48:841a/64 scope link 
  16.        valid_lft forever preferred_lft forever
  17. 16: br10: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
  18.     link/ether ee:df:66:ab:c2:4b brd ff:ff:ff:ff:ff:ff
  19.     inet6 fe80::ecdf:66ff:feab:c24b/64 scope link 
  20.        valid_lft forever preferred_lft forever
  21. 17: br20: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
  22.     link/ether 9e:4d:f4:0a:6d:13 brd ff:ff:ff:ff:ff:ff
  23.     inet6 fe80::9c4d:f4ff:fe0a:6d13/64 scope link 
  24.        valid_lft forever preferred_lft forever
  25. 18: vlan10@bond0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue master br10 state UP group default qlen 1000
  26.     link/ether ae:fd:60:48:84:1a brd ff:ff:ff:ff:ff:ff
  27. 19: vlan20@bond0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue master br20 state UP group default qlen 1000
  28.     link/ether ae:fd:60:48:84:1a brd ff:ff:ff:ff:ff:ff

查看创建的网桥

  1. root@server1:~# brctl show
  2. bridge name     bridge id               STP enabled     interfaces
  3. br10            8000.eedf66abc24b       no              vlan10
  4. br20            8000.9e4df40a6d13       no              vlan20

测试bond0 IP与外部网关连通性

  1. root@server1:~# ping 172.16.8.1 -c 3
  2. PING 172.16.8.1 (172.16.8.1)56(84) bytes of data.
  3. 64 bytes from 172.16.8.1: icmp_seq=1ttl=255time=1.55 ms
  4. 64 bytes from 172.16.8.1: icmp_seq=2ttl=255time=1.61 ms
  5. 64 bytes from 172.16.8.1: icmp_seq=3ttl=255time=1.62 ms
  7. --- 172.16.8.1 ping statistics ---
  8. 3 packets transmitted, 3 received, 0% packet loss, time 2004ms
  9. rtt min/avg/max/mdev =1.550/1.593/1.620/0.030 ms
  10. root@server1:~#

在server1安装kvm虚拟化环境，然后创建两个新的kvm网络，分别绑定到不同网桥

  1. cat >br10-network.xml<<EOF<network><name>br10-net</name><forwardmode="bridge"/><bridgename="br10"/></network>
  2. EOF
  3. cat >br20-network.xml<<EOF<network><name>br20-net</name><forwardmode="bridge"/><bridgename="br20"/></network>
  4. EOF
  6. virsh net-define br10-network.xml
  7. virsh net-define br20-network.xml
  8. virsh net-start br10-net
  9. virsh net-start br20-net
  10. virsh net-autostart br10-net
  11. virsh net-autostart br20-net

查看新建的网络

  1. root@server1:~# virsh net-list
  2.  Name       State    Autostart   Persistent
  3. ---------------------------------------------
  4.  br10-net   active   yesyes
  5.  br20-net   active   yesyes
  6.  default    active   yesyes

创建两个虚拟机，指定使用不同网络

  1. virt-install \--name vm1 \--vcpus1\--memory2048\--diskpath=/var/lib/libvirt/images/vm1/jammy-server-cloudimg-amd64.img \
  2.   --os-variant ubuntu22.04 \--import\--autostart\--noautoconsole\--networknetwork=br10-net
  4. virt-install \--name vm2 \--vcpus1\--memory2048\--diskpath=/var/lib/libvirt/images/vm2/jammy-server-cloudimg-amd64.img \
  5.   --os-variant ubuntu22.04 \--import\--autostart\--noautoconsole\--networknetwork=br20-net

查看创建的虚拟机

  1. root@server1:~# virsh list
  2.  Id   Name   State
  3. ----------------------
  4.  13   vm1    running
  5.  14   vm2    running

为vm1配置vlan10的IP地址

  1. virsh console vm1
  2. cat >/etc/netplan/00-installer-config.yaml<<EOF
  3. network:ethernets:enp1s0:addresses:- 172.16.10.10/24
  4.       nameservers:addresses:- 223.5.5.5
  5.       routes:-to: default
  6.         via: 172.16.10.1
  7.   version:2
  8. EOF
  9. netplan apply

为vm2配置vlan20的IP地址

  1. virsh console vm2
  2. cat >/etc/netplan/00-installer-config.yaml<<EOF
  3. network:ethernets:enp1s0:addresses:- 172.16.20.10/24
  4.       nameservers:addresses:- 223.5.5.5
  5.       routes:-to: default
  6.         via: 172.16.20.1
  7.   version:2
  8. EOF
  9. netplan apply

登录到vm1，测试vm1与外部网关连通性

  1. root@vm1:~# ip a1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
  2.     link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
  3.     inet 127.0.0.1/8 scope host lo
  4.        valid_lft forever preferred_lft forever
  5.     inet6 ::1/128 scope host 
  6.        valid_lft forever preferred_lft forever
  7. 2: enp1s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
  8.     link/ether 52:54:00:a4:aa:9d brd ff:ff:ff:ff:ff:ff
  9.     inet 172.16.10.10/24 brd 172.16.10.255 scope global enp1s0
  10.        valid_lft forever preferred_lft forever
  11.     inet6 fe80::5054:ff:fea4:aa9d/64 scope link 
  12.        valid_lft forever preferred_lft forever
  13. root@vm1:~# 
  14. root@vm1:~# ping 172.16.10.1 -c 3
  15. PING 172.16.10.1 (172.16.10.1)56(84) bytes of data.
  16. 64 bytes from 172.16.10.1: icmp_seq=1ttl=255time=1.51 ms
  17. 64 bytes from 172.16.10.1: icmp_seq=2ttl=255time=7.10 ms
  18. 64 bytes from 172.16.10.1: icmp_seq=3ttl=255time=2.10 ms
  20. --- 172.16.10.1 ping statistics ---
  21. 3 packets transmitted, 3 received, 0% packet loss, time 2003ms
  22. rtt min/avg/max/mdev =1.505/3.568/7.101/2.509 ms
  23. root@vm1:~#

登录到vm2，测试vm2与外部网关连通性

  1. root@vm2:~# ip a1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
  2.     link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
  3.     inet 127.0.0.1/8 scope host lo
  4.        valid_lft forever preferred_lft forever
  5.     inet6 ::1/128 scope host 
  6.        valid_lft forever preferred_lft forever
  7. 2: enp1s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
  8.     link/ether 52:54:00:89:61:da brd ff:ff:ff:ff:ff:ff
  9.     inet 172.16.20.10/24 brd 172.16.20.255 scope global enp1s0
  10.        valid_lft forever preferred_lft forever
  11.     inet6 fe80::5054:ff:fe89:61da/64 scope link 
  12.        valid_lft forever preferred_lft forever
  13. root@vm2:~# 
  14. root@vm2:~# ping 172.16.20.1 -c 3
  15. PING 172.16.20.1 (172.16.20.1)56(84) bytes of data.
  16. 64 bytes from 172.16.20.1: icmp_seq=1ttl=255time=1.73 ms
  17. 64 bytes from 172.16.20.1: icmp_seq=2ttl=255time=2.00 ms
  18. 64 bytes from 172.16.20.1: icmp_seq=3ttl=255time=2.00 ms
  20. --- 172.16.20.1 ping statistics ---
  21. 3 packets transmitted, 3 received, 0% packet loss, time 2003ms
  22. rtt min/avg/max/mdev =1.732/1.911/2.003/0.126 ms
  23. root@vm2:~#
标签： linux 网络 运维
本文转载自: https://blog.csdn.net/networken/article/details/137021517
版权归原作者 willops 所有， 如有侵权，请联系我们删除。
登录后发布评论

“linux 网卡配置 vlan/bond/bridge/macvlan/ipvlan/macvtap 模式”的评论:

还没有评论
关于作者
...
overfit同步小助手
文章同步
相关阅读

网络安全法-网络运行安全

使用selenium/drissionpage时如何阻止chrome自动跳转http到https

docker desktop 里部署的Open WebUI 管理员密码忘记了的处理方法

在ubuntu20.04中搭建onsite比赛运行环境

利用开源的低代码表单设计器FcDesigner高效管理和渲染复杂表单结构

Kafka学习笔记

【前端】浏览器输入url到页面呈现发生了什么？

文章导航
没有找到文章内容分段。。。