bgp bestpath as-path multipath-relax and IPv6

[jni]

Hello 

Quick question. 

I know bgp bestpath as-path multipath-relax is working for IPv4, but what about IPv6 route loadbalancing?

I did lab it in GNS3, but it didn't give the expected output for IPv6. 

For IPv4 before/after implementation of bgp bestpath as-path multipath-relax

#show ip bgp 0.0.0.0 
BGP routing tshow ip bgp 0.0.0.0 able entry for 0.0.0.0/0, version 21
Paths: (3 available, best #2, table Default-IP-Routing-Table)
Flag: 0x840
  Advertised to update-groups:
        1    2
  200, (received & used)
    200.200.200.1 from 200.200.200.1 (200.200.200.1)
      Origin IGP, metric 0, localpref 100, valid, external
  100, (received & used)
    100.100.100.1 from 100.100.100.1 (100.100.100.1)
      Origin IGP, metric 0, localpref 100, valid, external, best



#show ip bgp 0.0.0.0 
BGP routing table entry for 0.0.0.0/0, version 30
Paths: (3 available, best #2, table Default-IP-Routing-Table)
Multipath: eBGP
  Advertised to update-groups:
        1    2
  200, (received & used)
    200.200.200.1 from 200.200.200.1 (200.200.200.1)
      Origin IGP, metric 0, localpref 100, valid, external, multipath
  100, (received & used)
    100.100.100.1 from 100.100.100.1 (100.100.100.1)
      Origin IGP, metric 0, localpref 100, valid, external, multipath, best

But if I try the same for IPv6 I see no difference in output before/after

#sh bgp ipv6 unicast ::/0
BGP routing table entry for ::/0, version 2
Paths: (3 available, best #2, table Global-IPv6-Table)
  Advertised to update-groups:
        1    2
  200
    2200::1 (FE80::C005:9DFF:FEAF:1) from 2200::1 (200.200.200.1)
      Origin IGP, metric 0, localpref 100, valid, external
  100
    1100::1 (FE80::C004:9DFF:FE90:1) from 1100::1 (100.100.100.1)
      Origin IGP, metric 0, localpref 100, valid, external, best



#sh bgp ipv6 unicast ::/0
BGP routing table entry for ::/0, version 2
Paths: (2 available, best #2, table Global-IPv6-Table)
  Advertised to update-groups:
        1	2
  200
    2200::1 (FE80::C005:9DFF:FEAF:1) from 2200::1 (200.200.200.1)
      Origin IGP, metric 0, localpref 100, valid, external
  100
    1100::1 (FE80::C004:9DFF:FE90:1) from 1100::1 (100.100.100.1)
      Origin IGP, metric 0, localpref 100, valid, external, best

Is this a GNS3 issue, doesn't it work for IPv6 or is it just me :o)

Thank you in advance.

[Giuseppe Larosa · Accepted solution]

Hello @jni ,

can you post the configuration of your router , at least the router BGP section ?

Have you applied the command ar router bgp level or within an address family ?

ipv4 unicast

ipv6 unicast

It can also be a limitation of GNS3, but more info is needed to provide some help.

Hope to help

Giuseppe

[jni]

Since you where the only one to reply, and that you probably would have spottet the error when I posted the configuration, I accept your answer (but as I stated I didn't state maximum-paths under address-family ipv6)

[Giuseppe Larosa]

Hello @jni ,

when dealing with MP BGP most of the commands become per address family,

So my suggestion was a guess of what could be the most reasonable cause of the issue based on experience.

I'm happy to have been of help

Best Regards

Giuseppe

[jni · Accepted solution]

Argh... my own fault

I forgot to add maximum-paths under the ipv6 address-family. This of course has to be added under each address-family. My bad.