Help us improve your experience.

Let us know what you think.

Do you have time for a two-minute survey?

 
 

BNG Redundancy for DHCP Subscribers Using Packet Triggered Based Recovery

Learn about broadband network gateway (BNG) redundancy using packet triggered based recovery which provides simple, easy to use, and lightweight stateless subscriber redundancy.

BNG Redundancy for DHCP Subscribers Using Packet Triggered Based Recovery Overview

The BNG redundancy for DHCP subscribers using packet triggered based recovery provides simple, easy to use, and lightweight stateless redundancy with minimal traffic loss. The stateless BNG redundancy for DHCP subscribers supports dynamic C-VLAN and static VLAN model for both relay and server. The packet triggered based recovery utilizes the existing features such as auto configuration of VLAN and packet triggered subscribers.

Auto Configuration of VLAN

The auto configuration feature creates dynamic VLAN (DVLAN) logical interface on receiving the first VLAN packet from the client. On receiving the first packet, the Routing Engine authenticates the subscriber with authenticating server. The authentication server might need the accounting and advanced services details for authenticating the subscriber. The Routing Engine creates the DVLAN logical interface based on the request from the authenticating server. After creating the DVLAN logical interface, the system forwards the packet to the protocol stack for further processing.

Packet Triggered Subscribers

The packet-triggered subscriber feature creates IP demux logical interface on receiving a packet from clients with the pre-assigned IPv4 or IPv6 address. The forwarding plane validates the source IP address and matches with the configured IP address or prefix ranges. After the source IP address validation, the forwarding plane forwards the packet to the Routing Engine. The Routing Engine authenticates the subscriber with authenticating server as per the volume of accounting and advanced services such as firewall filter and CoS. Routing Engine creates IP demux logical interface as per the services requested by the authenticating server.

Benefits of BNG Redundancy for DHCP Subscribers Using Packet Triggered Based Recovery

  • Provides simple backup BNG deployment.

How BNG Redundancy for DHCP Subscribers Using Packet Triggered Based Recovery Works

Primary BNG hosts the subscribers during normal traffic flow. When the traffic flow fails in the primary BNG, the access nodes redirect the traffic to the backup BNG. The primary BNG can fail due to following reasons:

  • Intermediate node failure or link failure which breaks the MPLS path between access node and primary BNG.
  • Primary BNG link or port failure.
  • Primary BNG line card failure.
  • Primary BNG Routing Engine failure.
  • Primary BNG chassis failure.
  • Primary BNG to core network link failure.
Figure 1: L2 Circuit Based on IP/MPLS PWHT Scenario L2 Circuit Based on IP/MPLS PWHT Scenario

Figure 1 shows the topology diagram for layer 2 circuit based on IP/MPLS pseudowire headend termination (PWHT) scenario.

Based on the first traffic after failover, the Routing Engine creates DVLAN and dynamic IP subscriber. Packet Forwarding Engine forwards the subsequent traffic in the forwarding plane to the core router as per QoS and services attached to the IP subscriber. This QoS and the services are not the same QoS and services of the subscriber created in the Primary BNG. These are the common default dynamic IP subscriber profile features, assigned by RADIUS server or local configuration, until the session lease renewal and re-authentication occurs.

Once the system creates the DHCP subscriber in the secondary BNG, it provides limited QoS and other services at best-effort traffic with minimal interruption. When the DHCP client lease timer expires, it tries to re-negotiate lease time and a new DHCP protocol exchange takes place. This time, the system creates the fully functional DHCP subscriber along with QoS and advanced services as that of the primary BNG. The Packet Forwarding Engine forwards the traffic also to the core router accordingly. The system deletes the dynamic IP subscriber when the fully functional DHCP subscriber is active.

The traffic switchover to the backup BNG and the revert to the primary BNG process is similar. If revert occurs after the first lease timeout, the system proceeds with the switchover process. If revert occurs before the first lease timeout, the system proceeds with revert as it still has the previously assigned IP address and DHCP bindings.

The BNG redundancy using packet triggered based recovery feature supports the following access network topology for BNG redundancy:

  • Layer 2 VPN scenario
  • Layer 2 circuit based on IP/MPLS PWHT scenario
  • Ethernet VPN–virtual private wireless service (EVPN-VPWS).

Configuring BNG Redundancy Using Packet Triggered Recovery for DHCP Subscribers

Overview

Starting from 22.4R1, Junos supports Broadband Network Gateway (BNG) redundancy configuration using packet-trigger based recovery that provides easy to use and lightweight stateless subscriber redundancy.

This section provides a configuration example for packet triggered BNG redundancy for DHCP subscribers using a local DHCP server.

Requirements

The configuration example uses the following devices:

  • BNG1 and BNG2 broadband network gateways run EVPN-VPWS pseudowire headend termination (PWHT) with the ACX aggregation nodes and terminate the IPoE sessions. The BNG implements the packet triggered redundancy for the IPoE sessions.
  • ACX1 and ACX2 devices aggregate the Access Nodes traffic through the Cloud Metro Fabric towards the BNGs.
  • MX204 device is used for simulated peripheral connectivity.
  • vQFX instance for Q-in-Q tunneling and VLAN translation.

Topology

Figure 2 shows the physical topology with two vMX devices configured as BNG1 and BNG2 servers, two access devices ACX1 and ACX2, a vQFX and an MX204 device.

Figure 2: Topology Topology

Configuration

CLI Quick Configuration

Device BNG1

To quickly configure this example, copy the following commands, paste them into a text file, remove any line breaks, change any details necessary to match your network configuration, and then copy and paste the commands into the CLI at the [edit] hierarchy level.

Device BNG2

Device ACX1

Device ACX2

Step-by-Step Procedure

Configuring BNG1

  1. Log in to the BNG1 device. Ensure that the device is running Junos Release 22.4R1 or later versions.

  2. Configure system settings.

  3. Create a group to define common core interfaces configuration such as MTU, hold-time and damping parameters.

  4. Configure the interfaces towards core devices.

  5. Configure the interface towards the vQFX.

  6. Configure the loopback interface.

  7. Configure IS-IS protocol in the core network.

  8. Configure routing options.

  9. Configure the BGP protocol between the BNG and access devices

  10. Configure LDP and MPLS for all core interfaces.

  11. Configure global services for subscriber management, such as maintaining subscribers, tracing operations, and enabling enhanced subscriber management.

  12. Configure tracing options for the general authentication service.

  13. Configure system services, including tracing operations and Routing Engine failover, for the main enhanced subscriber management session management process, smg-service.

  14. Define global tracing operations for extended DHCP local server and extended DHCP relay agent processes.

  15. Configure the tunnel services and any additional chassis configuration.

  16. Configure access profiles for DHCP subscribers.

  17. Configure the pseudowire interface to use dynamic stacked VLANs. Also, configure additional interface and VLAN subscription settings. Configure Ethernet Segment Identifier (ESI) for EVPN active-standby multihoming.

  18. Configure internet VRF for internet routes.

  19. Configure the DHCP local server options on a routing instance. You will configure and apply routing policies for the DHCP subscriber routing instance, create domain maps, firewall filters and dynamic profiles for DHCP subscribers.

  20. Configure and apply dynamic profiles for pseudowire interface.

  21. Configure a routing instance of type evpn-vpws, defining route distinguisher and the VRF target.

Configuring BNG2

  1. Log in to the BNG2 device. Ensure that the device is running Junos Release 22.4R1 or later versions.

  2. Configure system services.

  3. Create a Group to define common core interfaces configuration such as MTU, hold-time and damping parameters.

  4. Configure the interfaces towards core devices.

  5. Configure the interface towards the vQFX.

  6. Configure the loopback interface for use in the subscriber management access network.

  7. Configure IS-IS protocol in the core network.

  8. Configure routing options.

  9. Configure the BGP protocol between the BNG and access devices.

  10. Configure LDP and MPLS for all core interfaces.

  11. Configure global services for subscriber management, such as maintaining subscribers, tracing operations, and enabling enhanced subscriber management.

  12. Configure tracing options for the general authentication service.

  13. Configure system services, including tracing operations and Routing Engine failover, for the main enhanced subscriber management session management process, smg-service.

  14. Define global tracing operations for extended DHCP local server and extended DHCP relay agent processes.

  15. Configure the tunnel services and any additional chassis configuration.

  16. Configure access profiles for DHCP subscribers.

  17. Configure the pseudowire interface to use dynamic stacked VLANs. Also, configure additional interface and VLAN subscription settings. Configure Ethernet Segment Identifier (ESI) for EVPN active-standby multihoming.

  18. Configure internet VRF for internet routes.

  19. Configure the DHCP local server options on a routing instance. You will configure and apply routing policies for the DHCP subscriber routing instance, create domain maps, firewall filters and dynamic profiles for DHCP subscribers.

  20. Configure and apply dynamic profiles for pseudowire interface.

  21. Configure a routing instance of type evpn-vpws, defining route distinguisher and the VRF target.

Configuring ACX1

  1. Log in to the ACX1 device.

  2. Configure System settings and DHCP service settings.

  3. Create a Group to define common core interfaces configuration such as MTU, hold-time and damping parameters.

  4. Configure the interfaces towards the core devices

  5. Configure aggregate interface with appropriate VLAN and Ethernet Segment Identifier (ESI) configuration.

  6. Configure the loopback interfaces.

  7. Configure the IS-IS protocol for the core network.

  8. Configure routing options.

  9. Configure the BGP protocol between the access devices and BNG.

  10. Configure the LDP and MPLS protocols on the core interfaces.

  11. Configure a routing instance of type evpn-vpws, defining route distinguisher and the VRF target.

Configuring ACX2

  1. Log in to the ACX2 device.

  2. Configure System Settings and DHCP service settings.

  3. Create a Group to define common core interfaces configuration such as MTU, hold-time and damping parameters.

  4. Configure the interfaces towards the core devices.

  5. Configure aggregate interface with appropriate VLAN and Ethernet Segment Identifier (ESI) configuration.

  6. Configure the loopback interfaces.

  7. Configure the IS-IS protocol for the core network.

  8. Configure routing options.

  9. Configure the BGP protocol between the access devices and BNG.

  10. Configure the LDP and MPLS protocols on the core interfaces.

  11. Configure a routing instance of type evpn-vpws, defining route distinguisher and the VRF target.

Verification

Purpose

You can verify the configuration by issuing the show evpn vpws-instance command on the BNG and access devices to view the details of the VPWS instance of the EVPN.

Action

  • Verifying on the BNG1 device.

  • Verifying on the BNG2 device.

  • Verifying on ACX1 device.

  • Verifying on the ACX2 device.

Meaning

The BNG devices are configured in an active-standby multihoming mode. In a steady state, all flows are directed towards BNG1, which is the primary device between BNG1 and BNG2. In case BNG1 encounters a failure, BNG2 becomes the primary designated forwarder. The access devices are configured in active-active multihoming, load balancing all the traffic from the CE devices.