（一）简述
1.双网卡绑定技术在centos7中使用了teaming技术，而在rhel6/centos7中使用的是bonding技术，
在centos7中双网卡绑定既能使用teaming也可以使用bonding，这里推荐使用teaming技术，方便与查看和监控。
- 原理
这里介绍两种最常见的双网卡绑定模式：
（1） roundrobin - 轮询模式
所有链路处于负载均衡状态，这种模式的特点增加了带宽，同时支持容错能力。
（2） activebackup - 主备模式
一个网卡处于活动状态，另一个处于备份状态，所有流量都在主链路上处理，当活动网卡down掉时，启用备份网卡。
3.查看服务器的相关信息：[[email protected]calhost ~]# nmcli dev DEVICE TYPE STATE CONNECTION em1 ethernet connected -- em2 ethernet connected -- em3 ethernet unavailable -- em4 ethernet unavailable -- lo loopback unmanaged -- [[email protected] ~]# ip a 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1 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: em1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq master team0 state UP qlen 1000 link/ether 36:3c:c4:19:d0:ef brd ff:ff:ff:ff:ff:ff 3: em2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq master team0 state UP qlen 1000 link/ether 36:3c:c4:19:d0:ef brd ff:ff:ff:ff:ff:ff 4: em3: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq state DOWN qlen 1000 link/ether 24:6e:96:a2:dd:ae brd ff:ff:ff:ff:ff:ff 5: em4: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc mq state DOWN qlen 1000 link/ether 24:6e:96:a2:dd:af brd ff:ff:ff:ff:ff:ff 18: team0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP qlen 1000 link/ether 36:3c:c4:19:d0:ef brd ff:ff:ff:ff:ff:ff inet 10.128.105.156/24 brd 10.128.105.255 scope global team0 valid_lft forever preferred_lft forever inet6 fe80::718:e575:f031:47d4/64 scope link valid_lft forever preferred_lft forever###########其中state UP 为网卡有网线或启用的网口，有上边的参数可知，em1和em2有网线接口，其他的网口没有。lo为本地回环网卡
（二）Roundrobin（轮询）安装步骤
1，查看网卡的连接信息：nmcli con sh
[[email protected] ~]# nmcli con sh
NAME UUID TYPE DEVICE
em1 4c8a8aee-327b-4f29-9ec4-ea2b2551420b 802-3-ethernet --
em2 62bde3ef-9f40-42aa-9b3a-4fe8b1aabf42 802-3-ethernet --
em3 56ef9d05-3a5d-4cd6-af5b-3d39adce619f 802-3-ethernet --
em4 761f7809-d10c-4e3b-bc06-c0906ca55d1c 802-3-ethernet -- 2，删除无用的网卡：nmcli con del em3
[[email protected] ~]# nmcli con del em3
Connection ‘em3‘ (56ef9d05-3a5d-4cd6-af5b-3d39adce619f) successfully deleted.
[[email protected] ~]# nmcli con sh
NAME UUID TYPE DEVICE
em1 4c8a8aee-327b-4f29-9ec4-ea2b2551420b 802-3-ethernet --
em2 62bde3ef-9f40-42aa-9b3a-4fe8b1aabf42 802-3-ethernet --
em4 761f7809-d10c-4e3b-bc06-c0906ca55d1c 802-3-ethernet -- 3，# 使用nmcli命令操作，创建team接口team0，同时设置teaming模式为roundrobin
[[email protected] ~]# nmcli con add type team con-name team0 ifname team0 config ‘{"runner":{"name": "roundrobin"}}‘
[[email protected] ~]# nmcli con sh
NAME UUID TYPE DEVICE
team0 5a1976d7-2c91-456e-8dda-0b12c6ef0c04 team team0
em1 4c8a8aee-327b-4f29-9ec4-ea2b2551420b 802-3-ethernet --
em2 62bde3ef-9f40-42aa-9b3a-4fe8b1aabf42 802-3-ethernet --
em4 761f7809-d10c-4e3b-bc06-c0906ca55d1c 802-3-ethernet -- 4，给接口team0设置ip地址和网关，并取消手动模式：
[[email protected] ~]# nmcli con modify team0 ipv4.address ‘10.128.105.157/24‘ ipv4.gateway ‘10.128.105.254‘
###设置为手动模式，取消DHCP
[[email protected] ~]# nmcli con modify team0 ipv4.method manual5，使用ifup team0 启用team0网卡，然后重启网络服务。
[[email protected] ~]# ifup team0
[[email protected] ~]# /etc/init.d/network restart
Restarting network (via systemctl): [ OK ]
[[email protected] ~]# ip a |grep team0
2: em1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq master team0 state UP qlen 1000
10: team0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP qlen 1000
inet 10.128.105.157/24 brd 10.128.105.255 scope global team06，再次将两张物理网卡添加到team
[[email protected] ~]# nmcli con add type team-slave con-name team0-port2 ifname em1 master team0
Connection ‘team0-port2‘ (e01c37f5-b3a1-4e6d-96ab-a6e676ad3c32) successfully added.
[[email protected] ~]# nmcli con add type team-slave con-name team0-port2 ifname em2 master team0
Connection ‘team0-port2‘ (8b13a9de-d64c-44fc-a86d-ea45d2f511cf) successfully added.
[[email protected] ~]# nmcli con sh
NAME UUID TYPE DEVICE
team0 64b123d5-189d-43b0-a850-77d47c0da86c team team0
team0-port1 24181898-bffe-4ae1-90e8-f47da64b5512 802-3-ethernet em1
team0-port2 8b13a9de-d64c-44fc-a86d-ea45d2f511cf 802-3-ethernet em2
em1 4c8a8aee-327b-4f29-9ec4-ea2b2551420b 802-3-ethernet --
em2 62bde3ef-9f40-42aa-9b3a-4fe8b1aabf42 802-3-ethernet --
em4 761f7809-d10c-4e3b-bc06-c0906ca55d1c 802-3-ethernet -- 7，现在查看team0的状态模式是否是roundrobin。teamdctl team0 st
[[email protected] ~]# teamdctl team0 st
setup:
runner: roundrobin
ports:
em1
link watches:
link summary: up
instance[link_watch_0]:
name: ethtool
link: up
down count: 0
em2
link watches:
link summary: up
instance[link_watch_0]:
name: ethtool
link: up
down count: 0这样，team模式的roundrobin搭建成功。
8，测试是否正常工作
8.1，首先把em1网卡禁用，用其他的服务器ping看是否断。
[[email protected] etc]# ifdown em1
Device ‘em1‘ successfully disconnected.
[[email protected] etc]# nmcli con sh
NAME UUID TYPE DEVICE
team0 64b123d5-189d-43b0-a850-77d47c0da86c team team0
team0-port2 8b13a9de-d64c-44fc-a86d-ea45d2f511cf 802-3-ethernet em2
em1 4c8a8aee-327b-4f29-9ec4-ea2b2551420b 802-3-ethernet --
em2 62bde3ef-9f40-42aa-9b3a-4fe8b1aabf42 802-3-ethernet --
em4 761f7809-d10c-4e3b-bc06-c0906ca55d1c 802-3-ethernet --
team0-port1 24181898-bffe-4ae1-90e8-f47da64b5512 802-3-ethernet --
[[email protected] ~]# teamdctl team0 st
setup:
runner: roundrobin
ports:
em2
link watches:
link summary: up
instance[link_watch_0]:
name: ethtool
link: up
down count: 0
从上图看出实在正常的。
8.2,把网卡em1启用，em2禁用查看下下[[email protected] etc]# ifup em1
[[email protected] etc]# nmcli con sh
NAME UUID TYPE DEVICE
team0 64b123d5-189d-43b0-a850-77d47c0da86c team team0
team0-port1 24181898-bffe-4ae1-90e8-f47da64b5512 802-3-ethernet em1
team0-port2 8b13a9de-d64c-44fc-a86d-ea45d2f511cf 802-3-ethernet em2
em1 4c8a8aee-327b-4f29-9ec4-ea2b2551420b 802-3-ethernet --
em2 62bde3ef-9f40-42aa-9b3a-4fe8b1aabf42 802-3-ethernet --
em4 761f7809-d10c-4e3b-bc06-c0906ca55d1c 802-3-ethernet --
[[email protected] etc]# ifdown em2
Device ‘em2‘ successfully disconnected.
[[email protected] etc]# nmcli con sh
NAME UUID TYPE DEVICE
team0 64b123d5-189d-43b0-a850-77d47c0da86c team team0
team0-port1 24181898-bffe-4ae1-90e8-f47da64b5512 802-3-ethernet em1
em1 4c8a8aee-327b-4f29-9ec4-ea2b2551420b 802-3-ethernet --
em2 62bde3ef-9f40-42aa-9b3a-4fe8b1aabf42 802-3-ethernet --
em4 761f7809-d10c-4e3b-bc06-c0906ca55d1c 802-3-ethernet --
team0-port2 8b13a9de-d64c-44fc-a86d-ea45d2f511cf 802-3-ethernet -- 网卡流量em2没变，而em1变化了。
（三）双网卡绑定activebackup的实现
具体的大概步骤和上边差不多，省略如下：
（首先删除刚才绑定的网卡模式轮询：nmcli con del team0 team0-port1 team0-port2）
1， 添加逻辑网卡team0，并启用activebackup模式[[email protected] ~]# nmcli con add type team con-name team0 ifname team0 config ‘{"runner":{"name":"activebackup"}}‘
2， 绑定ip
[[email protected] ~]# nmcli con modify team0 ipv4.address 10.128.105.157/24 ipv4.gateway 10.128.105.254
[[email protected] ~]# nmcli con modify team0 ipv4.method manual3， 添加物理网卡到activebackup模式
[[email protected] ~]# nmcli con add type team-slave con-name team0-port1 ifname em1 master team0
Connection ‘team0-port1‘ (24181898-bffe-4ae1-90e8-f47da64b5512) successfully added.
[[email protected] ~]# nmcli con add type team-slave con-name team0-port2 ifname em2 master team0
Connection ‘team0-port2‘ (8b13a9de-d64c-44fc-a86d-ea45d2f511cf) successfully added.4,查看状态
[[email protected] ~]# teamdctl team0 st
setup:
runner: activebackup
ports:
em1
link watches:
link summary: up
instance[link_watch_0]:
name: ethtool
link: up
down count: 0
em2
link watches:
link summary: up
instance[link_watch_0]:
name: ethtool
link: up
down count: 0
runner:
active port: em1成功，处于活动中的网卡为 eno33554984
备注：注意 在做网卡绑定的时候，如果发现物理网卡总是无法绑定到team0，请检查该物理网卡是否处于up状态。还有就是主备模式在切换的过程中会有间隔，会丢几个包。
原文地址：http://blog.51cto.com/liqingbiao/2089733
时间： 2024-08-01 22:43:03