We are currently testing a dual-stack setup on our BNG. IPv4 only is already productively running (XR 6.4.2).
The BNG plays DHCP server for both protocols, an external radius server assigns the attributes (ip, mask, vrf, interface for ipv4, delegated-prefix, ipaddrv6 for ipv6). All this seems to be fine, a session is created, the routing table gets populated and the CPE get his ipv4/ipv6 add with a prefix /62 subnet.
While the connectivity over ipv4 works seamlessly, ipv6 has some troubles. We are able to get it working by restarting the CPE and unplug the uplink for approximately two minutes and the client behind the CPE is able to reach the internet over ipv6. If we trigger another restart of the CPE, the bng session for ipv4 and ipv6 get released and reestablished after a few minutes, everything seems fine but the client is not able to reach the internet over ipv6.
We'd appreciate if someone can share their experience with a BNG dual stack setup, especially in case of ipv6 if there's something to look for. Let me know if if can provide some configuration snippets to give you a better insight.
there are many dual-stack deployments, but as there are hardly two BNG deployments that are alike, it would be the best if you could open a service request with Cisco TAC to have this issue properly investigated.
Crosswork Cloud - Crosswork Traffic Analysis - FAQ
Crosswork Cloud - Crosswork Traffic Analysis is a Cloud-hosted Software as a Service platform that provides Netflow based Traffic Analytics. The Crosswork Traffic Analysis platform Traffic Analysis, Peeri...
Cisco Champion Radio · S8|E9 Innovations to Achieve a Trustworthy Infrastructure
How do you know for certain that a router in your network has not been altered with since you deployed it? Wouldn’t it be great if you can cryptographically challenge your r...
IOS upgrade on asr9xx mandates rommon upgrades sometimes while they can be optional at other times. You may land up in unwanted situation if proper procedure is not followed during upgrades.
This article will include complete details about rommon ...
In some situation NCS560 RP become unresponsive after reload or powercycle.
In many NCS560 deployments are in remote location, deployment might be large and human intervention should be kept at minimum
Engineering team have been working on a str...
In simple terms, 'Route Churn' is defined as the 'rate of change of prefixes'. Different XR versions across 4.x to 7.x have differing behavior & support for the BGP churn handling and some enhancements made from 6.5.3 onwards (listed in appendix) mak...