IPv6 BGP network advertisement

  • 0
  • 1
  • Problem
  • Updated 2 years ago
  • Solved
Hi Experts

Getting error while trying to advertise IPv6 routes in BGP peering. IPv4 works fine, but getting error in IPv6. The below network is preset in IPv6 routing table and shows as being directly connected.

(vr Test) SwitchA # configure bgp add network address-family ipv6-unicast fd01:976a:14fd:e7c0::/68

Error: vr Test: IPv6 network is not valid, must be a global unicast address and not a unique local address

Photo of Sumanta Ghosh

Sumanta Ghosh

  • 458 Points 250 badge 2x thumb

Posted 2 years ago

  • 0
  • 1
Photo of Sumanta Ghosh

Sumanta Ghosh

  • 458 Points 250 badge 2x thumb
Photo of Frank

Frank

  • 3,662 Points 3k badge 2x thumb
Your netmask, /68, is that a typo?
Because I think that /64 is the smallest size you can advertise via routing protocols.

I'm also not sure what happens when you use private V6 networks (the ones starting with fd) and put them in BGP. Part of me says it should work, but I've never tried it. Just like I've never tried putting "192.168.1.0" into BGP.

   Frank
Photo of Nick Yakimenko

Nick Yakimenko

  • 2,404 Points 2k badge 2x thumb
well, if it is IGP -- why not? BGP announcements between internal routers
Photo of Henrique

Henrique, Employee

  • 10,302 Points 10k badge 2x thumb
Hi Sumanta, based on the error message you should add a Global Unicast Address.

GUA is currently defined by IPv6 prefix 2000::/3 which gives you the range below:

           2000:0000:0000:0000:0000:0000:0000:0000 to 3fff:ffff:ffff:ffff:ffff:ffff:ffff:ffff

ULA is currently defined by IPv6 prefix FC00::/7 which gives you the range below:

           fc00:0000:0000:0000:0000:0000:0000:0000 to fdff:ffff:ffff:ffff:ffff:ffff:ffff:ffff
Photo of Grosjean, Stephane

Grosjean, Stephane, Employee

  • 12,582 Points 10k badge 2x thumb
what version of EXOS are you running, btw?
Photo of Sumanta Ghosh

Sumanta Ghosh

  • 458 Points 250 badge 2x thumb
Hi

Please see below:-

ExtremeXOS version 21.1.2.14
Photo of Edward Tsui

Edward Tsui, Employee

  • 602 Points 500 badge 2x thumb
This is clearly stated in RFC4193:

The default behavior of exterior routing protocol sessions between administrative routing regions must be to ignore receipt of and not advertise prefixes in the FC00::/7 block.  A network operator may specifically configure prefixes longer than FC00::/7 for inter-site communication.
If BGP is being used at the site border with an ISP, the default BGP configuration must filter out any Local IPv6 address prefixes, both incoming and outgoing.  It must be set both to keep any Local IPv6 address prefixes from being advertised outside of the site as well as to keep these prefixes from being learned from another site.  The exception to this is if there are specific /48 or longer routes created for one or more Local IPv6 prefixes.
Photo of Sumanta Ghosh

Sumanta Ghosh

  • 458 Points 250 badge 2x thumb
Hi Ed

So is this default behavior as per RFC? Extreme doesn't allow LUA to be used in BGP advertisement, even if it's within internal network (not globally routable) ?
Photo of Henrique

Henrique, Employee

  • 10,302 Points 10k badge 2x thumb
Hi Sumanta, we are in touch with our engineering team regarding this topic and will update this thread as soon as we have more details.

We know (based on the RFC) that there are some restrictions when exchanging ULA prefixed between sites. However, since we are talking about iBGP, my guess is that it should work.