Troubleshooting PPPoE BNG on Telecom Routers: Configuration, Compatibility & Error Resolving

Overview & Thematic Scope

Deploying a Broadband Network Gateway (BNG) with PPPoE termination on telecom routers is a critical step for ISPs delivering subscriber access. This FAQ addresses the most common technical and deployment questions engineers face during setup, from session scaling and RADIUS integration to troubleshooting PPPoE negotiation failures and optimizing forwarding performance.

Frequently Asked Questions

Q1: What is the recommended baseline configuration for PPPoE BNG on a telecom router?

A baseline PPPoE BNG configuration requires defining a virtual-template interface, binding it to a physical or logical Ethernet interface, and enabling PPPoE session termination. You must also configure an IP address pool, RADIUS or local authentication, and a default route toward the core. For scale, enable per-session QoS and DHCPv6-PD if IPv6 subscribers are supported.

Q2: How many simultaneous PPPoE sessions can a typical telecom router support as a BNG?

Session capacity depends on the router’s control-plane CPU, memory, and forwarding ASIC. Mid-range telecom routers typically support 8,000 to 64,000 PPPoE sessions, while high-end chassis can scale to 256,000 or more. Always verify the vendor’s session-scale specification with features like QoS, ACLs, and IPv6 enabled, as these reduce maximum session counts.

Q3: Why do PPPoE sessions fail to establish with error ‘PADI timeout’ or ‘PADO not received’?

‘PADI timeout’ usually indicates a Layer 2 connectivity or VLAN mismatch between the subscriber and the BNG. Check that the access interface is up, the correct VLAN is tagged, and PPPoE discovery is not blocked by a firewall or MAC filter. On the BNG side, confirm the virtual-template is bound to the correct interface and that no static MAC limits are exhausted.

Q4: How do I integrate RADIUS authentication and accounting for PPPoE BNG subscribers?

Configure the router as a RADIUS client with the server IP, shared secret, and authentication/accounting ports (typically 1812/1813). Bind the RADIUS server group to the PPPoE virtual-template and enable accounting updates. Ensure the RADIUS attributes for IP address assignment, session timeout, and QoS profiles are supported by both the router and the RADIUS server.

Q5: What MTU and MRU settings should I use for PPPoE BNG to avoid fragmentation?

Set the PPPoE MTU and MRU to 1492 bytes to account for the 8-byte PPPoE header within a 1500-byte Ethernet frame. If the subscriber uses PPPoE over VLAN or QinQ, subtract additional tag bytes (4 per tag) from the MTU. Mismatched MTU/MRU is a common cause of large-packet drops and slow throughput.

Q6: How can I troubleshoot high CPU usage on a PPPoE BNG router?

High CPU is often caused by excessive PPPoE session churn, control-plane policing not enabled, or per-packet logging. Enable control-plane policing for PPPoE discovery and PPP LCP/NCP packets, disable debug logging in production, and consider hardware-based session offload if supported. Also check for RADIUS retransmissions or unreachable servers causing queue buildup.

Q7: Is PPPoE BNG compatible with IPv6 and dual-stack subscriber deployments?

Yes, most modern telecom routers support PPPoE with IPv6CP and DHCPv6-PD for dual-stack. You must enable IPv6 on the virtual-template, configure an IPv6 address pool or prefix delegation pool, and ensure RADIUS returns the correct IPv6 attributes. Verify that the forwarding ASIC supports IPv6 PPPoE session termination at the required scale.

Q8: What are the best practices for high availability and redundancy in PPPoE BNG deployments?

Implement BNG redundancy using inter-chassis session redundancy (ICSR) or VRRP with PPPoE session synchronization. Subscriber sessions should be synchronized between active and standby BNGs so failover preserves IP addresses and session state. Test failover regularly and ensure RADIUS and DHCP servers are reachable from both nodes to minimize MTTR.