nmcli实现Network Teaming

Posted 2021-11-25

网络组：是将多个网卡聚合在一起方法，从而实现冗错和提高吞吐量
网络组不同于旧版中bonding技术，提供更好的性能和扩展性
网络组由内核驱动和teamd守护进程实现.
多种方式runner
broadcast
roundrobin
activebackup
loadbalance
lacp (implements the 802.3ad Link Aggregation Control Protocol)

这是已经配好的，NetworkManger支持多配置文件，想删时需要先down网卡

[[email protected] ~]# nmcli connection show
[[email protected] ~]# nmcli connection down NWteam0

[[email protected] ~]# nmcli connection delete NWteam0 (删除网卡配置文件)

1创建网络组接口

nmcli con add type team con-name CNAME ifname INAME [config JSON]
CNAME 连接名，INAME 接口名
JSON 指定runner方式
格式：‘{"runner": {"name": "METHOD"}}‘
METHOD 可以是broadcast, roundrobin,
activebackup, loadbalance, lacp

[[email protected] ~]# nmcli connection add con-name OFFICE type team ifname office config ‘{"runner":{"name":"loadbalance"}}‘ ipv4.addresses 192.168.153.150 ipv4.method manual

2创建port接口

[[email protected] ~]# nmcli connection add con-name OFFICE-1 type team-slave ifname ens33 master OFFICE
[[email protected] ~]# nmcli connection add con-name OFFICE-2 type team-slave ifname ens38 master OFFICE

(绿色代表已开启的配置文件)

3测试

[[email protected] ~]# nmcli connection up OFFICE # 开启OFFICE配置