Link aggregation based on team mechanism

Preface

stay linux in , Rhel7 It used to be bond Mechanism To bind multiple network cards to the same IP Address , To provide access to the network , And according to different modes Load balancing perhaps Recurrent succession management Processing data . And by the Rhel7 after , Provides a powerful tool , nmcli Tool command , Use this tool , The specific configuration file will be regenerated according to the configuration of command parameters for use by the network interface , Convenient and flexible . stay Rhel7 in , No longer use bond Mechanism definition , Instead, use netgroups team Mechanism , hold team Group as a device

Example

Next, let's do a simple experiment

---

First add two network cards

View the network card and its binding information ：nmcli device ,nmcli connection

Obviously ,ens33 It's our original network card , And successfully connected , and ens37 and ens38 It's the network card we just added , We can configure it manually
Now? , We need to check Team Sample file of the network group and select the link aggregation model
route ：/usr/share/doc/teamd/example_configs/, We cd Go and have a look

We choose the first file , And intercept part of it ：cat activebackup_arp_ping_1.conf

Here's the screenshot ："runner": {"name": "activebackup"}（ There is no comma ）

add to Team Device and session ：nmcli connection add ifname team0 con-name team0 type team config '{"runner": {"name": "activebackup"}}'

We configure it manually IP Address （ It's OK here, too dhcp Automatic access to ）：nmcli connection modify team0 ipv4.method manual ipv4.addresses 192.168.32.168/24 ipv4.gateway 192.168.32.1 ipv4.dns 114.114.114.114

We have successfully matched it with 168 The address of （ I'll match whatever I want ）
Add device to team0 ：（team0-port1 I wrote it casually ）

1. nmcli connection add type team-slave ifname ens37 con-name team0-port1 master team0
2. nmcli connection add type team-slave ifname ens38 con-name team0-port2 master team0

Activate the slave device ：nmcli connection up team0-port1
nmcli connection team0-port2

Activate the master device ：nmcli connection up team0

Check the status ：teamdctl team0 state

You can see , Currently active devices are ens37
We use another device ping once 168（ This machine ）

At this time , We are close ens37：nmcli connection down team0-port1

At this time, the only thing left to work is ens38, Let's see if the other host can still ping through

Obviously , Still can , This is the advantage of link aggregation

---

The last step , Delete Team0
disconnect ：nmcli connection down team0
Delete team0：nmcli connection delete team0
nmcli connection delete team0-port1
nmcli connection delet team0-port2