I had no idea how convoluted VLANs could get until I tried to implement them in netlab.

We’ll start with the simplest option: a single VLAN stretched across two bridges switches with two Linux hosts connected to it. netlab can configure VLANs on Arista EOS, Cisco IOSv, Cisco Nexus OS, VyOS, Dell OS10, and Nokia SR Linux. We’ll use the quickest (deployment-wise) option: Arista EOS on containerlab.

We’ll use groups in the lab topology file to define our devices. Members of the hosts group will be Linux containers, members of the switches group will be Arista EOS containers using vlan configuration module:

provider: clab

nodes: [ h1, h2, s1, s2 ]

groups:
  hosts:
    members: [ h1, h2 ]
    device: linux
  switches:
    members: [ s1, s2 ]
    module: [ vlan ]
    device: eos

Next step: defining VLANs. All we have to do is to define the VLAN name. 802.1q tag and VXLAN VNI are assigned automatically.

vlans:
  red:

Finally, we have to specify the links in our lab. We’ll set the access VLAN name with vlan.access attribute. On the switch-to-host links, we have to specify the access VLAN on the switch interface. On the inter-switch link we can set vlan.access as a link attribute – it will be propagated to all switches connected to the link.

links:
- h1:
  s1:
    vlan.access: red
- s1:
  s2:
  vlan.access: red
- s2:
    vlan.access: red
  h2:

VLAN configuration on s1 and s2 is performed automatically as part of the netlab up process. Here’s the VLAN-related configuration from s1:

vlan 1000
   name red
!
interface Ethernet1
   mac-address 52:dc:ca:fe:03:01
   switchport access vlan 1000
!
interface Ethernet2
   mac-address 52:dc:ca:fe:03:02
   switchport access vlan 1000
!
interface Vlan1000
   description VLAN red (1000) -> [h1,s2,h2]
   ip address 172.16.0.3/24

Here’s the corresponding configuration on Cisco IOSv¹:

bridge irb
!
interface GigabitEthernet0/1
 bridge-group 1
!
interface GigabitEthernet0/2
 bridge-group 1
!
interface BVI1
 description VLAN red (1000) -> [h1,s2,h2]
 ip address 172.16.0.3 255.255.255.0
!
bridge 1 protocol ieee
bridge 1 route ip

VLAN IP Addressing

A netlab VLAN segment is a single subnet; the IPv4 and IPv6 prefixes are allocated from the lan address pool. All physical links and VLAN interfaces belonging to the same access VLAN use the same IP subnet. The IPv4 addresses in our lab are thus set up as follows:

The Linux hosts (OK, containers) can ping the IPv4 addresses assigned to the switch VLAN interfaces, but not the loopback IP addresses of both switches – the Linux hosts have a default route pointing to one of the switches, and there’s no routing configured between the switches:

$ netlab connect h1
Connecting to container clab-vlan-access-stretch-h1, starting bash
h1:/# ping -c 5 -q h2
PING h2 (172.16.0.2): 56 data bytes
--- h2 ping statistics ---
5 packets transmitted, 5 packets received, 0% packet loss
round-trip min/avg/max = 3.261/3.475/3.598 ms
h1:/# ping -c 5 -q s1
PING s1 (10.0.0.3): 56 data bytes
--- s1 ping statistics ---
5 packets transmitted, 5 packets received, 0% packet loss
round-trip min/avg/max = 1.616/1.734/1.823 ms
h1:/# ping -c 5 -q s2
PING s2 (10.0.0.4): 56 data bytes
--- s2 ping statistics ---
5 packets transmitted, 0 packets received, 100% packet loss
h1:/# ping -c 5 -q 172.16.0.4
PING 172.16.0.4 (172.16.0.4): 56 data bytes
--- 172.16.0.4 ping statistics ---
5 packets transmitted, 5 packets received, 0% packet loss
round-trip min/avg/max = 3.040/3.232/3.363 ms

Next time, we’ll fix IP routing and add OSPF routing process to the switches. You could also do it on your own:

• Set up a Ubuntu VM or a bare-metal server with netlab, Docker and containerlab
• Install Arista cEOS container image
• Copy the lab topology file into an empty directory
• Start the lab with netlab up and start exploring