Skip to content

eve-setup.md

Latest commit

 

History

History
108 lines (84 loc) · 3.84 KB

eve-setup.md

File metadata and controls

108 lines (84 loc) · 3.84 KB
Raw

EVE-NG setup (VMware Workstation + Cisco 7206VXR / c7200)

1. Host prep (Windows 11)

  • Enable Intel VT-x / AMD-V in UEFI.
  • VMware Workstation 17.x. If nodes boot slowly, reduce Windows hypervisor contention: Core Isolation / Memory integrity OFF, and (optional, breaks WSL2) bcdedit /set hypervisorlaunchtype off + reboot.

2. EVE-NG VM

  • Import the EVE-NG Community OVA.
  • VM spec: 4 vCPU, 8 GB RAM, 60 GB disk.
  • VM Settings → Processors → check Virtualize Intel VT-x/EPT (VM powered off).
  • Network adapters:
    • NIC 1 → pnet0 = EVE management. Bridged (or NAT). Used for the web UI, SSH to EVE, and the dashboard container.
    • NIC 2 → pnet1 = lab OOB. Attach to a dedicated host-only VMnet (e.g. VMnet2, DHCP off). Becomes Cloud1 inside EVE and carries the router MGMT VRF subnet 192.168.99.0/24.
  • First boot: console as root / eve, set password, hostname, static mgmt IP, DNS, then apt update && apt -y upgrade, reboot.

3. Give the EVE host an IP on the MGMT bridge

So the container (and routers' default route) have a gateway at 192.168.99.1:

# /etc/network/interfaces.d/pnet1  (or edit /etc/network/interfaces)
auto pnet1
iface pnet1 inet static
    address 192.168.99.1
    netmask 255.255.255.0
    bridge_ports eth1
    bridge_stp off

ifreload -a (or reboot). Confirm ip addr show pnet1 → 192.168.99.1/24.

4. Cisco 7206VXR (c7200 / Dynamips) image

/opt/unetlab/addons/dynamips/c7200-adventerprisek9-mz.152-4.S6.image
/opt/unetlab/wrappers/unl_wrapper -a fixpermissions

No license file needed. If your image name differs, set it in inventory.yaml (lab.image) or pass --image to build_lab.py.

Node profile used by the generated lab: NPE-400, 256 MB RAM (512 MB makes the emulated CPU halt), PA-8E in slot 1 → Fa0/0 (mgmt) + Ethernet1/0 .. 1/7. The EVE c7200 template only offers PA-FE-TX / PA-4E / PA-8E. 8 nodes ≈ 2 GB.

5. Build the lab (generate + import)

python backend/scripts/build_lab.py        # writes labs/bgp-attributes.unl

In EVE-NG: Main page → Import and upload labs/bgp-attributes.unl (or copy it into /opt/unetlab/labs/ and unl_wrapper -a fixpermissions). The file defines all 8 c7200 nodes, the 12 point-to-point links and the Cloud1 (pnet1) management bridge — no manual wiring.

Then Start all nodes. On first boot, per node: right-click → Idle PC and pick a value with a low CPU marker (the lab ships 0x606df838; adjust if your host CPU sits high). Lab settings → enable Startup-configuration.

6. One-time SSH bring-up (over the console)

Fresh c7200 nodes have no user, no RSA key, no MGMT IP. Bring the dashboard container up first (step below), then:

docker exec -it bgp-attributes-executor python scripts/bootstrap.py

It telnets each node's EVE-NG console port and configures the local user, domain name, RSA key, SSH-only VTY, and Fa0/0 in VRF MGMT. After that every node answers SSH on its 192.168.99.x address and you can push full configs:

docker exec -it bgp-attributes-executor python scripts/push_baseline.py
docker exec -it bgp-attributes-executor python scripts/healthcheck.py

7. Verify the path before touching the dashboard

ping 192.168.99.111                 # from the EVE host
ssh lab@192.168.99.111              # EDGE1
#   show ip ospf neighbor
#   show ip bgp summary
#   show ip bgp 100.100.100.0/24   # expect a path via ISP-A and via iBGP

8. Optional — put the dashboard behind EVE's nginx

/etc/nginx/sites-enabled/ add:

location /bgp/ {
    proxy_pass http://127.0.0.1:8000/;
    proxy_set_header Host $host;
    proxy_http_version 1.1;
    proxy_set_header Connection "";        # keep SSE alive
    proxy_buffering off;
}

nginx -t && systemctl reload nginx → dashboard at http://<eve>/bgp/.