溫和的方式請參考：https://www.cnblogs.com/zzf0305/p/9594093.html
一：傳統的bond方式(飯已驗證)------------本種的綁定方式比較暴躁
（1） bond幾種模式介紹。
mode 0 ：load balancing（round-robin）模式，需要交換機端支持，支持多端口負載均衡，支持端口冗余，slave接口的mac相同
mode 1：active-backup模式，最大支持兩個端口，一主一備，同一時間只有一塊網卡工作，不支持搶占
mode 4:采用IEEE802.3ad方式的動態協商機制聚合端口，需要交換機開啟lacp並配置為主動（active）模式
mode 5和 mode 6類似mode 1的主備模式，不常用
（2）bond配置
需要關閉NetworkManager服務
systemctl stop NetworkManager systemctl disable NetworkManager
查看內核是否加載bonding
lsmod |grep bonding (如果未加載，用#modprobe --first-time bonding 飯：這只是臨時的加載，重啟就沒了，永久的需要做下面的配置文件) 配置bonding驅動 vi /etc/modprobe.d/bond.conf (沒有則新建)，寫入以下內容： alias bond0 bonding options bond0 miimon=100 mode=0 //miimon是用來進行鏈路監測的，后面指定的是檢查的間隔時間，單位是ms
注：網卡配置文件，“=”左邊均為大寫，右邊為小寫，如bond0的不成功，請細心檢查配置文件。
（3）配置bond接口
# vi /etc/sysconfig/network-scripts/ifcfg-bond0(新建，寫入以下內容)
TYPE=Bond
BOOTPROTO=none
ONBOOT=yes
USERCTL=no //是否允許普通用戶控制此設備
DEVICE=bond0
IPADDR=192.168.0.111
PREFIX=24
NM_CONTROLLED=no //NetworkManager服務的參數，配置修改后無重啟立即生效
BONDING_MASTER=yes
我的實際配置文件：
[root@cnbgdphapwanp01 network-scripts]# cat ifcfg-bond0 DEVICE="bond0" BOOTPROTO=none ONBOOT="yes" IPADDR=10.11.1.137 NETMASK=255.255.255.0 GATEWAY=10.11.1.1 TYPE=Ethernet
（4）配置slave接口
vi /etc/sysconfig/network-scripts/ifcfg-ens33 TYPE=Ethernet BOOTPROTO=none NAME=ens33 DEVICE=ens33 ONBOOT=yes MASTER=bond0 SLAVE=yes USERCTL=no
其他slave網卡與此配置相同
我的實際配置文件
[root@cnbgdphapwanp01 network-scripts]# cat ifcfg-bond1 DEVICE="bond1" BOOTPROTO=none ONBOOT="yes" IPADDR=221.99.229.226 NETMASK=255.255.255.0 GATEWAY=211.99.229.21 TYPE=Ethernet
（5）重啟network服務，並檢查
systemctl restart network cat /proc/net/bonding/bond0
二.NetworkManager服務的nmcli方式（暫時未測試）
推薦參考自己的另一篇：https://www.cnblogs.com/zzf0305/p/9594093.html
（1）查看網絡設備狀態
[root@compute ~]# nmcli dev DEVICE TYPE STATE CONNECTION eth0 ethernet connected eth0 eth1 ethernet connected Wired connection 1 lo loopback unmanaged --
（2）查看網絡連接狀態
[root@compute ~]# nmcli con sh NAME UUID TYPE DEVICE Wired connection 1 d75d7715-1098-353e-bb11-4b718e51ff38 802-3-ethernet eth1 eth0 5fb06bd0-0bb0-7ffb-45f1-d6edd65f3e03 802-3-ethernet eth0
（3）創建team0(也就是bond接口）
按照下面的語法，用 nmcli 命令為網絡組接口創建一個連接。
nmcli con add type team con-name CNAME ifname INAME [config JSON]
CNAME 指代連接的名稱，INAME 是接口名稱，JSON (JavaScript Object Notation) 指定所使用的處理器(runner)。JSON語法格式如下：
‘{"runner":{"name":"METHOD"}}‘
METHOD 是以下的其中一個：broadcast、activebackup、roundrobin、loadbalance 或者 lacp。
下面以“roundrobin”為例：
[root@compute ~]# nmcli con add type team con-name team0 ifname team0 config ‘{"runner":{"name":"roundrobin"}}‘
Connection ‘team0‘ (64021ca5-85c3-429d-b930-56802dc0ccc4) successfully added.
設置team0的ip,gateway,dns
[root@compute ~]# nmcli con modify team0 ipv4.address "192.168.0.222/16" ipv4.gateway "192.168.0.1" [root@compute ~]# nmcli con modify team0 ipv4.dns "223.5.5.5"
設置team0的屬性為手動（manual）
[root@compute ~]# nmcli con modify team0 ipv4.method manual
添加slave網卡
[root@compute ~]# nmcli con add type team-slave con-name team-port2 ifname eth1 master team0 Connection ‘team-port2‘ (df74a4c7-f8ff-4ae3-b04f-3dd1210598cd) successfully added. [root@compute ~]# nmcli con add type team-slave con-name team-port1 ifname eth0 master team0 Connection ‘team-port1‘ (757648c4-114f-439f-b022-5bcf63ae0cb3) successfully added.
啟動team0網口，並檢查
[root@compute ~]# nmcli con up team0
Connection successfully activated (master waiting for slaves) (D-Bus active path: /org/freedesktop/NetworkManager/ActiveConnection/3)
[root@compute ~]# teamdctl team0 sta
setup:
runner: roundrobin
ports:
eth0
link watches:
link summary: up
instance[link_watch_0]:
name: ethtool
link: up
down count: 0
eth1
link watches:
link summary: up
instance[link_watch_0]:
name: ethtool
link: up
down count: 0
三.清除bond信息(未測試)
1. #
ifconfig bond0 down
2. 清除對應配置文件信息
3. #lsmod |grep bonding 存在則用rmmod bonding刪除，更改NetworkManager開機自啟，重啟系統。
查看網卡速率： ethtool 網卡名（eth0）
常見故障：
啟動team0網口，team0仍舊為down
[root@compute network-scripts]# 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: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:4f:fd:82 brd ff:ff:ff:ff:ff:ff
inet 192.168.0.222/16 brd 192.168.255.255 scope global eth0
valid_lft forever preferred_lft forever
inet6 fe80::20c:29ff:fe4f:fd82/64 scope link
valid_lft forever preferred_lft forever
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:4f:fd:8c brd ff:ff:ff:ff:ff:ff
inet 192.168.0.160/16 brd 192.168.255.255 scope global dynamic eth1
valid_lft 7191sec preferred_lft 7191sec
inet 192.168.0.159/16 brd 192.168.255.255 scope global secondary dynamic eth1
valid_lft 6300sec preferred_lft 6300sec
inet6 fe80::86d1:12d7:5a7c:2d88/64 scope link
valid_lft forever preferred_lft forever
4: team0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN
排錯：
1.檢查網絡連接狀態，發現team-port1和team-port2以及team0沒有連接到網卡設備
[root@compute network-scripts]# nmcli con sh
NAME UUID TYPE DEVICE
eth0 5fb06bd0-0bb0-7ffb-45f1-d6edd65f3e03 802-3-ethernet eth0
eth1 22a287d8-6206-4d10-bdd9-5299b063300e 802-3-ethernet eth1
team-port1 757648c4-114f-439f-b022-5bcf63ae0cb3 802-3-ethernet --
team-port2 df74a4c7-f8ff-4ae3-b04f-3dd1210598cd 802-3-ethernet --
team0 64021ca5-85c3-429d-b930-56802dc0ccc4 team --
2.刪除eth0和eth1的連接
[root@compute network-scripts]# nmcli con del eth0 eth1
3再次查看發現team0及slave接口正常連接到設備
[root@compute ~]# nmcli con sh
NAME UUID TYPE DEVICE
team-port1 757648c4-114f-439f-b022-5bcf63ae0cb3 802-3-ethernet eth0
team-port2 df74a4c7-f8ff-4ae3-b04f-3dd1210598cd 802-3-ethernet eth1
team0 64021ca5-85c3-429d-b930-56802dc0ccc4 team team0
4.查看team0接口狀態並測試連通性
[root@compute ~]# 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: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master team0 state UP qlen 1000
link/ether 00:0c:29:4f:fd:82 brd ff:ff:ff:ff:ff:ff
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast master team0 state UP qlen 1000
link/ether 00:0c:29:4f:fd:82 brd ff:ff:ff:ff:ff:ff
4: team0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP qlen 1000
link/ether 00:0c:29:4f:fd:82 brd ff:ff:ff:ff:ff:ff
inet 192.168.0.222/16 brd 192.168.255.255 scope global team0
valid_lft forever preferred_lft forever
inet6 fe80::ac47:e724:cd16:c5ca/64 scope link tentative dadfailed
valid_lft forever preferred_lft forever
inet6 fe80::acce:9394:eafe:57bb/64 scope link tentative dadfailed
valid_lft forever preferred_lft forever
inet6 fe80::e1a2:77fd:6148:c7c6/64 scope link tentative dadfailed
valid_lft forever preferred_lft forever
[root@compute ~]# ping baidu.com
PING baidu.com (111.13.101.208) 56(84) bytes of data.
64 bytes from 111.13.101.208 (111.13.101.208): icmp_seq=1 ttl=52 time=30.6 ms
64 bytes from 111.13.101.208 (111.13.101.208): icmp_seq=1 ttl=52 time=30.7 ms (DUP!)
^C
--- baidu.com ping statistics ---
1 packets transmitted, 1 received, +1 duplicates, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 30.684/30.696/30.708/0.012 ms
[root@compute ~]#
注意測試中出現以下狀況是由於交換機端沒有做端口聚合配置造成
[root@compute ~]# ping baidu.com
PING baidu.com (111.13.101.208) 56(84) bytes of data.
64 bytes from 111.13.101.208 (111.13.101.208): icmp_seq=1 ttl=52 time=28.2 ms
64 bytes from 111.13.101.208 (111.13.101.208): icmp_seq=1 ttl=52 time=28.2 ms (DUP!)
64 bytes from 111.13.101.208 (111.13.101.208): icmp_seq=2 ttl=52 time=29.2 ms
64 bytes from 111.13.101.208 (111.13.101.208): icmp_seq=2 ttl=52 time=29.2 ms (DUP!)
64 bytes from 111.13.101.208 (111.13.101.208): icmp_seq=3 ttl=52 time=29.8 ms
64 bytes from 111.13.101.208 (111.13.101.208): icmp_seq=3 ttl=52 time=29.9 ms (DUP!)
64 bytes from 111.13.101.208 (111.13.101.208): icmp_seq=4 ttl=52 time=27.7 ms
64 bytes from 111.13.101.208 (111.13.101.208): icmp_seq=4 ttl=52 time=27.7 ms (DUP
轉自：©著作權歸作者所有：來自51CTO博客作者BigManer的原創作品，如需轉載，請注明出處，否則將追究法律責任