[j-nsp] bgp ipv6 route problem

[Kazmierczak, Tomasz (NSN - US/Irving)]

Somehow I thought we re talking here about eBGP session... But it looks U re trying
configure inet6 labeled-unicast between these 2x peers if U re getting ::FFFF:<ipv4> as a next hop for ipv6 routes.
Is this your intention? Trying to configure 6PE? 

If just regular ipv6 bgp on ipv4 session, thei idea is the same, U need to configure a dual stack ipv4/ipv6 addresses on interfaces between these 2x routers, for example:
ge-1/3/2 {
    unit 0 {
        family inet {
            address 10.48.0.58/30;
        }
        family inet6 {
            address ::10.48.0.58/126;
        }
    }
And define family inet/inet6 in bgp, for example:

protocols {
    bgp {
        group ibgp {
            type internal;
            family inet {
                unicast;
            }
            family inet6 {
                unicast;
            }
            neighbor 10.48.0.57;
        }
    }

and U will get your ipv6 routes..
inet6.0: 8 destinations, 9 routes (8 active, 0 holddown, 0 hidden)
* 7::/64 (1 entry, 1 announced)
     Accepted
     Nexthop: ::10.48.0.57
     Localpref: 100
     AS path: 1 2 3 I

If U re playing with 6PE U still have more stuff to do..

---
-Tomek


-----Original Message-----
From: Kazmierczak, Tomasz (NSN - US/Irving)
Sent: Fri 3/16/2012 12:10 PM
To: ext bruno; Per Granath; juniper-nsp
Subject: RE: [j-nsp] bgp ipv6 route problem
 
Hello,

I am running 10.4 and still next hop is in format ::x.y.z.v, this was not changed.

With regards to your orig questions, there are 2x ways of doing it,
for example make eBGP peering using p2p IPv4 addresses were Ipv4 and compaIPv4/Ipv6 address is defined. In this case U just need define family ipv4/ipv6 under your bgp group and you will be able see routes on other box.

Brgds
Tomasz

-----Original Message-----
From: juniper-nsp-bounces at puck.nether.net [mailto:juniper-nsp-bounces at puck.nether.net] On Behalf Of ext bruno
Sent: Friday, March 16, 2012 10:52 AM
To: Per Granath; juniper-nsp
Subject: Re: [j-nsp] bgp ipv6 route problem

Hello per,


Thanks for ur response.R1 dont accept the route in 10.4. how do we do make it accept the route.


Thanks
 


 
 
------------------ Original ------------------
From:  "Per Granath"<per.granath at gcc.com.cy>;
Date:  Fri, Mar 16, 2012 11:46 PM
To:  "juniper-nsp"<juniper-nsp at puck.nether.net>; 
Cc:  "bruno"<bruno.juniper at gmail.com>; 
Subject:  RE: [j-nsp] bgp ipv6 route problem

 
In 10.4 the automatically created "IPv4-compatible IPv6-address" changed, that is the (::ffff:12.1.1.1). 
Before 10.4 it used to be just (::12.1.1.1).

If you have mixture in the network it will be confusing...

BTW, the JNCIE exam is now using 10.4.

> i met some odd problem on junos 10.4 . two router(R1----R2) establish ipv4
> bgp session with ipv4 and ipv6 family. session is up. only the ipv4 route is
> accept. not ipv6 route.  so i trace on r1 . i find the ipv6 route is ignore.    but
> not problem on junos 9.4.
> lab at junos104.# run show bgp summary logical-system r1
> 12.1.1.1              400         26         28       0       1           9 Establ
>   inet.0: 1/1/1/0
>   inet6.0: 0/0/0/0
> 
> 
> 
> Mar 15 03:12:42.988838 bgp_nexthop_sanity: peer 12.1.1.1 (External AS 600)
> next hop ::ffff:12.1.1.1 unexpectedly remote, ignoring routes in this update
> Mar 15 03:12:42.988853 bgp_rcv_nlri: Peer 12.1.1.1 (External AS 600) Mar 15
> 03:12:42.988865 bgp_rcv_nlri: 2222:2222:2222::/48 Mar 15 03:12:42.988880
_______________________________________________
juniper-nsp mailing list juniper-nsp at puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp