Skip to content

veesix osvbng#

veesix osvbng node is identified with veesix_osvbng kind in the topology file.

The integration of veesix osvbng has been tested with v0.3.1 release. Note, that releases before v0.3.1 are not supported and will not work with containerlab.

Getting veesix osvbng image#

The veesix osvbng container image is available publicly on Docker Hub as veesixnetworks/osvbng.

docker pull veesixnetworks/osvbng:<tag>

To pull the latest available image:

docker pull veesixnetworks/osvbng:latest

Managing veesix osvbng nodes#

The osvbng Northbound API is running on port 8080 by default. An OpenAPI Swagger UI is available at http://<node-name>:8080/api/docs/.

curl http://<node-name>:8080/api/show/protocols/ospf/neighbors

SSH server is running on the management interface.

ssh admin@<container-name>

The osvbngcli utility provides a basic interactive CLI for managing and monitoring the osvbng node:

docker exec -it <container-name> osvbngcli

osvbng> show subscriber sessions

Interface naming#

veesix osvbng container uses the following mapping for its interfaces:

  • eth0 - management interface connected to the containerlab management network
  • eth1+ - data interfaces

Features and options#

Startup configuration#

veesix osvbng uses a YAML configuration file to define interfaces, subscriber groups, DHCP settings, AAA policies, and routing protocols.

With the startup-config property of the node/kind, a user sets the path to the local config file that will be mounted to the container at /etc/osvbng/osvbng.yaml.

topology:
  nodes:
    bng1:
      kind: veesix_osvbng
      image: veesixnetworks/osvbng:v0.3.1
      startup-config: bng1/osvbng.yaml

Environment variables#

Variable Description Default
OSVBNG_WAIT_FOR_INTERFACES Wait for interfaces to be provisioned before starting (set automatically by containerlab) true

Quickstart#

The osvbng01 lab example demonstrates a minimal BNG topology with an instance of BNG Blaster as a subscriber simulator, an osvbng node, and a core router running FRR, giving you the ability to simulate a real-world BNG with QinQ IPoE termination.

The topology consists of three nodes:

  • subscribers - a BNG Blaster container simulating subscribers with Q-in-Q tagged IPoE sessions over DHCPv4/DHCPv6
  • bng1 - the osvbng node performing subscriber termination
  • corerouter1 - an FRR router acting as the core/upstream router
name: osvbng01

topology:
  nodes:
    bng1:
      kind: veesix_osvbng
      image: veesixnetworks/osvbng:v0.3.1
      startup-config: bng1/osvbng.yaml
    corerouter1:
      kind: linux
      image: frrouting/frr:v8.4.1
      binds:
        - corerouter1/daemons:/etc/frr/daemons
        - corerouter1/frr.conf:/etc/frr/frr.conf
    subscribers:
      kind: linux
      image: veesixnetworks/bngblaster:0.9.30
      binds:
        - subscribers/config.json:/config/config.json

  links:
    - endpoints: ["subscribers:eth1", "bng1:eth1"]
    - endpoints: ["bng1:eth2", "corerouter1:eth1"]