As part of my new duties as a SD-WAN engineer, I need to be proficient at understanding SD-WAN, OMP, TLOCs, etc. Because of this, I dropped the $200 necessary on Cisco CML to gain access to the SD-WAN images. Here I am documenting my progress installing Cisco SD-WAN on my EVE-NG server (R730). I have been primarily following the networkacademy.io guide (a bit outdated) and Nischal Prajapati’s blog for installing and standing this up. Screenshots provided as I remember to take them:

Github Files: https://github.com/Lenniusce/sd-wan

Images used on EVE

IP Scheme
VPN 0 – Used for the transport network (underlay)

VPN 512 – Used for management traffic. Interfaces in VPN 512 provide out-of-band access for management services such as SSH, HTTPS, SNMP, and communication with vManage.

System IP (System-ID) – A unique identifier for each SD-WAN device. It is similar to a router ID and is used to identify the device within the SD-WAN fabric

First, upload all of the files listed above to opt/unetlab/addon/qemu folder based on version name as show.

#  Creating Folders 
mkdir /opt/unetlab/addons/qemu/vtbond-20.15.1
mkdir /opt/unetlab/addons/qemu/vtsmart-20.15.1
mkdir /opt/unetlab/addons/qemu/vtmgmt-20.15.1
mkdir /opt/unetlab/addons/qemu/c8000v-17.15.01
mkdir /opt/unetlab/addons/qemu/vios-adventerprisek9-m.SPA.159-3.M4
mkdir /opt/unetlab/addons/qemu/linux-ubuntu-22.04-desktop

Remember, when creating the node images that we need to give 100GB to vManage in the form of a separate disk with the command:

/opt/qemu/bin/qemu-img create -f qcow2 virtiob.qcow2 100G

Fix permissions for all images once they are all uploaded:

/opt/unetlab/wrappers/unl_wrapper -a fixpermissions

Then, we need to create a new lab for SD-WAN in EVE-NG and provide a name, author, etc.

All of the viptela and Ubuntu images should show up on menu when building the topology now

Note: Viptela was the company Cisco acquired in 2017 to gain its SD-WAN technology. Cisco’s SD-WAN solution is built on the original Viptela platform, which is why many image names, controllers (vManage, vSmart, vBond), and software packages still use the “v” prefix and “viptela” naming. Even though the product is now branded Cisco SD-WAN, the underlying software heritage remains Viptela.

Topology is this so far, I have named paths (switches) based on their bandwidth (Low, Med, High, Direct):

Now, lets bootup vManage by right clicking on the node and selecting “Start”

Login to vManage, “admin/admin” are the default credentials

You will go thorough a mini wizard:

Changed Password to “Cisco”

Select 1 “COMPUTE AND DATA”

Select the 100gb drive, which should be “vdb”

After vmanage is running, start uploading the “initial” configs into the devices, I am following the IP Scheme from the top table. Telenet into each device and use Guacamole to copy-paste the configs into the devices (Ctrl + Shift)

vManage config

conf t
!
system
  host-name vManage
  system-ip 1.1.1.20
  site-id 1
  organization-name Chickenhawk
  clock timezone America/Los_Angeles
  vbond 10.1.1.10
vpn0
  interface eth0
  ip address 10.1.1.20/24
  ipv6 dhcp-client
  Tunnel-interface
  allow-service all
  allow-service dhcp
  allow-service dns
  allow-service icmp
  allow-service https
  exit
  no shutdown
  exit
  ip route 0.0.0.0/0 10.1.1.1
  exit
vpn 512
  interface eth1
  ip address 192.168.115.20/24
  no shutdown
!
exit
!
commit

vBond config

conf t
!
system
  host-name vBond
  system-ip 1.1.1.10
  site-id 1
  organization-name Chickenhawk
!
  clock timezone America/Los_Angeles
  vbond 10.1.1.10 local
!
vpn 0
  interface ge0/0
  ip address 10.1.1.10/24
  no tunnel-interface
  no shutdown
!
exit
!
commit

vSmart config

conf t
!
system
  host-name vSmart
  system-ip 1.1.1.30
  site-id 1
  organization-name Chickenhawk
!
  Clock timezone America/Los_Angeles
  vbond 10.1.1.10
!
vpn 0
  interface eth0
  ip address 10.1.1.30/24
  no tunnel-interface
  no shutdown
!
exit
!
commit

Modem_Next_Hop Config (simulating the endpoint)

Interface GigabitEthernet0/0
ip address 10.1.1.1 255.255.255.0
Interface GigabitEthernet0/1
ip address 10.1.10.1 255.255.255.0
Interface GigabitEthernet0/2
ip address 10.1.20.1 255.255.255.0
Interface GigabitEthernet0/3
ip address 10.1.50.1 255.255.255.0

Now that we have all of the configs in place, we should be able to ping across the entire topology using the 10.1.1.X addresses. If you cannot reach a particular box, shut/no shut the interface as I have seen this a few times with EVE-NG.

We are using the Ubuntu linux server as the GUI and SSH access to each device.

Set the IP address per the table above:

(This is the EVE-NG provided image, so its “Cisco/Test123”)

Hint: Make sure the “Direct” switch in between them is started, because I definitely forgot to boot it and couldn’t reach:

Access the SD-WAN vManage Console from the Firefox browser in the Ubuntu image at https://192.168.115.20:8443 it should be booted up by now

Next, check out my post on certificates as I pick right up where I left off: https://router-cake.com/2026/06/21/sd-wan-on-my-eve-ng-bare-metal-server-certificates/