netlab EVPN/VXLAN Bridging Example

netlab release 1.3 introduced support for VXLAN transport with static ingress replication and EVPN control plane. Last week we replaced a VLAN trunk with VXLAN transport, now we’ll replace static ingress replication with EVPN control plane.

Lab topology

Lab topology

We’ll start with the VXLAN bridging topology and add EVPN configuration module to the switches. We’ll also have to add BGP configuration module (EVPN needs BGP) and define BGP AS with bgp.as global parameter… and that’s it – now we have a working EVPN/VXLAN lab.

Adding EVPN and BGP configuration modules
groups:
  switches:
    members: [ s1,s2 ]
    module: [ vlan,vxlan,ospf,bgp,evpn ]

bgp.as: 65000

Behind the scenes netlab:

  • As in the previous example, configures OSPF and VXLAN. VXLAN VNIs no longer use the ingress replication lists.
  • Starts BGP process on S1 and S2, and builds an IBGP session between the loopback interfaces.
  • Assigns EVPN RT/RD to every VLAN with a VXLAN VNI.
  • Activates EVPN address family on the IBGP session between S1 and S2.

Now for an interesting plot twist. Everything netlab does works on multiple network operating systems1. All you have to do to build a multi-vendor lab using Arista EOS and Cumulus Linux is to add device types to the switches.

Building a multi-vendor EVPN/VXLAN lab
nodes:
  s1:
    device: eos
  s2:
    device: cumulus

Start the lab with netlab up and it all works. Here’s the BGP table as observed on S1 (Arista EOS):

EVPN BGP table on Arista EOS
s1#sh bgp evpn
BGP routing table information for VRF default
Router identifier 10.0.0.5, local AS number 65000
Route status codes: s - suppressed, * - valid, > - active, E - ECMP head, e - ECMP
                    S - Stale, c - Contributing to ECMP, b - backup
                    % - Pending BGP convergence
Origin codes: i - IGP, e - EGP, ? - incomplete
AS Path Attributes: Or-ID - Originator ID, C-LST - Cluster List, LL Nexthop - Link Local Nexthop

          Network                Next Hop              Metric  LocPref Weight  Path
 * >     RD: 10.0.0.6:1000 mac-ip aac1.ab17.1b03 fe80::a8c1:abff:fe17:1b03
                                 10.0.0.6              -       100     0       i
 * >     RD: 10.0.0.6:1001 mac-ip aac1.ab17.1b03 fe80::a8c1:abff:fe17:1b03
                                 10.0.0.6              -       100     0       i
 * >     RD: 10.0.0.6:1000 mac-ip aac1.abcf.8cf5
                                 10.0.0.6              -       100     0       i
 * >     RD: 10.0.0.6:1000 mac-ip aac1.abcf.8cf5 172.16.0.2
                                 10.0.0.6              -       100     0       i
 * >     RD: 10.0.0.6:1000 imet 10.0.0.6
                                 10.0.0.6              -       100     0       i
 * >     RD: 10.0.0.6:1001 imet 10.0.0.6
                                 10.0.0.6              -       100     0       i
 * >     RD: 10.0.0.5:1000 imet 10.0.0.5
                                 -                     -       -       0       i
 * >     RD: 10.0.0.5:1001 imet 10.0.0.5
                                 -                     -       -       0       i

You can also use netlab to learn how to configure netlab-supported features on platforms you’re not familiar with2. Here are the relevant snippets of Arista EOS configuration generated during the netlab initial process:

Arista cEOS EVPN/VXLAN bridging configuration
hostname s1
!
spanning-tree mode mstp
!
vlan 1000
   name red
!
vlan 1001
   name blue
!
interface Ethernet1
   switchport access vlan 1000
!
interface Ethernet2
   switchport access vlan 1001
!
interface Ethernet3
   description s1 -> s2
   no switchport
   ip address 10.1.0.1/30
   ip ospf network point-to-point
   ip ospf area 0.0.0.0
!
interface Loopback0
   ip address 10.0.0.5/32
   ip ospf area 0.0.0.0
!
interface Vlan1000
   description VLAN red (1000) -> [h1,h2,s2]
!
interface Vlan1001
   description VLAN blue (1001) -> [h3,h4,s2]
!
interface Vxlan1
   vxlan source-interface Loopback0
   vxlan udp-port 4789
   vxlan vlan 1000 vni 101000
   vxlan vlan 1001 vni 101001
!
router bgp 65000
   router-id 10.0.0.5
   no bgp default ipv4-unicast
   bgp advertise-inactive
   neighbor 10.0.0.6 remote-as 65000
   neighbor 10.0.0.6 next-hop-self
   neighbor 10.0.0.6 update-source Loopback0
   neighbor 10.0.0.6 description s2
   neighbor 10.0.0.6 send-community standard extended
   !
   vlan 1000
      rd 10.0.0.5:1000
      route-target import 65000:1000
      route-target export 65000:1000
      redistribute learned
   !
   vlan 1001
      rd 10.0.0.5:1001
      route-target import 65000:1001
      route-target export 65000:1001
      redistribute learned
   !
   address-family evpn
      neighbor 10.0.0.6 activate
   !
   address-family ipv4
      neighbor 10.0.0.6 activate
      network 10.0.0.5/32
!
router ospf 1
   router-id 10.0.0.5
   max-lsa 12000
!
end

Collecting Cumulus Linux configuration is a bit more of an effort. Here’s a summary of how the bridging interfaces are configured (collected from a variety of /etc/network/interfaces.d files):

Cumulus Linux interface configuration
iface bridge
	bridge-vlan-aware yes
	bridge-vids 1001
	bridge-vids 1000
	bridge-ports swp1
	bridge-ports swp2
	bridge-ports vni-101000
	bridge-ports vni-101001

iface swp1
	bridge-access 1000
	
iface swp2
	bridge-access 1001
	
iface vni-101000
	bridge-access 1000
	vxlan-id 101000
	vxlan-learning no
	
iface vni-101001
	bridge-access 1001
	vxlan-id 101001
	vxlan-learning no

… and here is the FRR control plane configuration:

Cumulus Linux FRR configuration
router bgp 65000
 bgp router-id 10.0.0.6
 no bgp default ipv4-unicast
 neighbor 10.0.0.5 remote-as 65000
 neighbor 10.0.0.5 description s1
 neighbor 10.0.0.5 update-source lo
 !
 address-family ipv4 unicast
  network 10.0.0.6/32
  neighbor 10.0.0.5 activate
  neighbor 10.0.0.5 next-hop-self
 exit-address-family
 !
 address-family l2vpn evpn
  neighbor 10.0.0.5 activate
  advertise-all-vni
  vni 101001
   rd 10.0.0.6:1001
   route-target import 65000:1001
   route-target export 65000:1001
  exit-vni
  vni 101000
   rd 10.0.0.6:1000
   route-target import 65000:1000
   route-target export 65000:1000
  exit-vni
  advertise-svi-ip
  advertise ipv4 unicast
 exit-address-family
!
router ospf
 ospf router-id 10.0.0.6

Want to run this lab on your own, or try it out with different devices? No problem:

  1. netlab supports 15 different platforms, and most configuration modules support at least a half-dozen platforms. ↩︎

  2. You’ll get working configurations that might not be optimal or adhering to vendor best practices. If you feel we did a bad job configuring a platform you’re an expert in, please submit a pull request. ↩︎

Recent posts in the same categories

Add comment
Home
Sidebar