Centos7网卡绑定的两种方法

https://blog.51cto.com/youdong/1963416

选择模式4，并且layer选择3+4，交换机要配置LAG

一．传统的bond方式

（1）bond几种主要模式介绍

ü mode 0

load balancing (round-robin)模式 ，需要交换机端支持，支持多端口负载均衡，支持端口冗余，slave接口的mac相同

ü mode 1

active-backup模式，最大支持两个端口，一主一备，同一时间只有一块网卡工作，不支持抢占

ü mode 4

采用IEEE802.3ad方式的动态协商机制聚合端口，需要交换机开启lacp,并且配置为主动（active）模式

ü mode5和mode6

类似mode1的主备模式，不常用

（2）bond配置

ü 需要关闭NetworkManager服务

[root@docker ~]# systemctl stop NetworkManager
[root@docker ~]# systemctl disable NetworkManager

 

ü 查看内核是否加载bounding

[root@docker ~]# lsmod|grep bonding
bonding               141566  0

 

如果没有加载bonding可以通过以下命令加载

modprobe --first-time bonding

ü 配置bonding驱动

[root@docker ~]# vi /etc/modprobe.d/bond.conf  //没有的话手动创建
[root@docker ~]# cat /etc/modprobe.d/bond.conf
alias bond0 binding
options bond0 miimon=100 mode=0   //miimon是用来进行链路监测的,后面指定的是检查的间隔时间，单位是ms

 

ü 配置bond接口

cat /etc/sysconfig/network-scripts/ifcfg-bond0
TYPE=Bond
BOOTPROTO=none
ONBOOT=yes
USERCTL=no   //USERCTL:是否允许普通用户控制此设备
DEVICE=bond0
IPADDR=192.168.0.221
PREFIX=24
NM_CONTROLLED=no //NetworkManger服务的参数，配置修改后无重启立即生效
BONDING_MASTER=yes

 

ü 配置slave接口

cat /etc/sysconfig/network-scripts/ifcfg-eth0
TYPE=Ethernet
BOOTPROTO=none
DEFROUTE=yes
PEERDNS=yes
PEERROUTES=yes
NAME=eth0
DEVICE=eth0
ONBOOT=yes
MASTER=bond0
SLAVE=yes
USERCTL=no

 

其他slave网卡与此配置相同

ü 重启network服务，并检查

[root@docker network-scripts]# systemctl restart network
[root@docker network-scripts]# cat /proc/net/bonding/bond0
Ethernet Channel Bonding Driver: v3.7.1 (April 27, 2011)

Bonding Mode: load balancing (round-robin)
MII Status: up
MII Polling Interval (ms): 0
Up Delay (ms): 0
Down Delay (ms): 0

Slave Interface: eth0
MII Status: up
Speed: 1000 Mbps
Duplex: full
Link Failure Count: 0
Permanent HW addr: 00:0c:29:ff:31:80
Slave queue ID: 0

Slave Interface: eth1
MII Status: up
Speed: 1000 Mbps
Duplex: full
Link Failure Count: 0
Permanent HW addr: 00:0c:29:ff:31:8a
Slave queue ID: 0
[root@docker network-scripts]#

 

二．NetworkManager服务的nmcli方式

（1）查看网络设备状态

[root@compute ~]# nmcli dev
DEVICE  TYPE      STATE      CONNECTION
eth0    ethernet  connected  eth0
eth1    ethernet  connected  Wired connection 1
lo      loopback  unmanaged  --
[root@compute ~]#

 

（2）查看网络连接状态

[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
[root@compute ~]#

 

（3）创建team0(也就是bond接口）

按照下面的语法，用 nmcli 命令为网络组接口创建一个连接。

# nmcli con add type team con-name CNAME ifname INAME [config JSON]

CNAME 指代连接的名称，INAME 是接口名称，JSON (JavaScript Object Notation) 指定所使用的处理器(runner)。JSON语法格式如下：

'{"runner":{"name":"METHOD"}}'

METHOD 是以下的其中一个：broadcast、activebackup、roundrobin、loadbalance 或者 lacp。

下面以“roundrobin”为例：

[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.
[root@compute ~]#

 

设置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"
[root@compute ~]#

 

设置team0的属性为手动（manual）

[root@compute ~]# nmcli con modify team0 ipv4.method manual

 

添加slave网卡

[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.
[root@compute ~]#

 

启动team0网口，并检查

[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
[root@compute ~]#

 

常见故障：

启动team0网口，team0仍旧为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

 

排错：

1.检查网络连接状态，发现team-port1和team-port2以及team0没有连接到网卡设备

 [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.删除eth0和eth1的连接

[root@compute network-scripts]# nmcli con del eth0 eth1

 

3再次查看发现team0及slave接口正常连接到设备

[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接口状态并测试连通性

[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 ~]#

 

注意测试中出现以下状况是由于交换机端没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