Arista EOS lab 10 of 20
VXLAN and EVPN with an MLAG leaf pair
Dual-home an access switch to an MLAG leaf pair that acts as one VTEP: shared VTEP loopback, iBGP underlay between the peers, EVPN to the spine, anycast gateways with the MLAG system MAC, and a failover test that keeps traffic flowing.
Run this lab
It needs Standard: its 7 nodes are set to 7 vCPU and 11,264 MB in total.
In your workspace: Shared, then Arista EOS guided labs, then 10 - VXLAN and EVPN with an MLAG leaf pair
- Level
- Advanced
- Time
- About 90 minutes
- Nodes need
- 7 vCPU, 11,264 MB
- Smallest plan
- Standard
The scenario
Northwind's web servers must survive a leaf failure, so rack 1 now has an MLAG pair, leaf1a and leaf1b, with the web access switch acc1 dual-homed to both. The pair's MLAG domain (POD1) and its eBGP sessions to spine1 are already up. Rack 2's leaf2 is a fully configured EVPN leaf hosting the database (VLAN 20, VRF TENANT-A, from lab 9), and VLAN 10 is already stretched to it for the web tier's future rack-2 servers. Your job is to make the MLAG pair a single EVPN VTEP, so the web tier in rack 1 reaches the database in rack 2 and keeps working when either rack-1 leaf fails. (One spine keeps the lab within the Standard plan; production would use two or more.)
| Device | Loopback0 | Loopback1 (VTEP) | AS | Notes |
|---|---|---|---|---|
| spine1 | 10.255.0.1 | – | 65000 | route server for EVPN |
| leaf1a | 10.255.0.11 | 10.255.1.11 (shared) | 65001 | Vlan4094 10.255.255.1/30 · Vlan4093 10.255.254.1/30 |
| leaf1b | 10.255.0.12 | 10.255.1.11 (shared) | 65001 | Vlan4094 10.255.255.2/30 · Vlan4093 10.255.254.2/30 |
| leaf2 | 10.255.0.13 | 10.255.1.13 | 65002 | VLAN 20 DB, gateway 10.20.20.1; VLAN 10 WEB is stretched here too (anycast gateway 10.10.10.1, no hosts yet) |
| h1 | – | – | – | on acc1 Et3, VLAN 10, 10.10.10.11/24 via 10.10.10.1 |
| h2 | – | – | – | on leaf2 Et3, VLAN 20, 10.20.20.12/24 via 10.20.20.1 |
Overlay values: VLAN 10 ↔ VNI 10010, RT 10:10010, RD <Loopback0>:10010; VRF TENANT-A ↔ VNI 50001, RT 50001:50001, RD <Loopback0>:50001; anycast MAC 00:1c:73:00:00:99.
What you’ll practise
- Dual-home a downstream switch to an MLAG pair with an MLAG port-channel
- Give both MLAG peers the same VTEP address so the pair is one VTEP to the rest of the fabric
- Add an iBGP underlay session between the peers with next-hop-self, and explain why it's needed
- Use the MLAG system ID as the VXLAN router MAC for routed traffic
- Verify with show vxlan config-sanity and from a remote leaf's view of the pair
- Compare MLAG dual-homing with EVPN all-active multihoming (ESI, route types 1 and 4)
The tasks
- Dual-home acc1
- iBGP between the MLAG peers
- One VTEP for two switches
- EVPN overlay and tenant on the pair
- Test, then fail a leaf link
The Lab Tasks pane has the scenario, these tasks, the checks to run, hints and the full solution.
Topology and devices
The network as it opens in EVE-NG, drawn from the lab’s own canvas, and the image each node runs.
Every link, as text
- spine1
Ethernet1to leaf1aEthernet1 - spine1
Ethernet2to leaf1bEthernet1 - spine1
Ethernet3to leaf2Ethernet1 - leaf1a
Ethernet2to leaf1bEthernet2 - leaf1a
Ethernet3to leaf1bEthernet3 - leaf1a
Ethernet4to acc1Ethernet1 - leaf1b
Ethernet4to acc1Ethernet2 - acc1
Ethernet3to h1eth0 - leaf2
Ethernet3to h2eth0
| Node | Device | Image in Add Node | Set to |
|---|---|---|---|
| spine1 | Arista vEOS | veos-4.33.1.1F | 1 vCPU, 2,048 MB |
| leaf1a | Arista vEOS | veos-4.33.1.1F | 1 vCPU, 2,048 MB |
| leaf1b | Arista vEOS | veos-4.33.1.1F | 1 vCPU, 2,048 MB |
| leaf2 | Arista vEOS | veos-4.33.1.1F | 1 vCPU, 2,048 MB |
| acc1 | Arista vEOS | veos-4.33.1.1F | 1 vCPU, 2,048 MB |
| h1 | Alpine Linux | clbalpinebase-alpine-base-3-20-3 | 1 vCPU, 512 MB |
| h2 | Alpine Linux | clbalpinebase-alpine-base-3-20-3 | 1 vCPU, 512 MB |
How to open this lab
Sign in to your workspace at
lab.cloudlabbox.com.In File Manager, open Shared, then the folder
Arista EOS guided labs.Open
10 - VXLAN and EVPN with an MLAG leaf pairand start its nodes, then open Lab Tasks for the scenario and the steps.
The lab is read-only. You can start the nodes and configure the devices, but changes to the topology can’t be saved. To change it, rebuild it in your own folder.
More Arista EOS labs
Related labs in other courses
CloudLabBox is independent of Arista Networks. The lab follows exam topics and isn’t vendor training material; vendor names identify the exams and the images in the workspace.