Le 2017-10-13 21:30, Paolo Lucente a écrit :
> Hi Fabien,
>
> There was no progress on the OP (that is, choose an IPv4 and an IPv6
> address to listen to; but you can choose an IPv6 or an IPv6 address).
>
> Any chance you can make a tcpdump of the BGP session and send it over
> via unicast email? I have the scenario working perfectly for me (that
> is, v4 transport travelling both v4 and v6 AFs). Do you have cases in
> which you have a v4 NLRI with a v6 BGP next-hop (*)? That should work
> too but may have been a less tested scenario.
>
> Cheers,
> Paolo
>
> (*) https://tools.ietf.org/html/rfc5549
>
> On Wed, Oct 11, 2017 at 10:43:49PM +0200, Fabien VINCENT wrote: Hi List ;)
>
> Paolo any news on this request ? Checking with IPv4/IPv6 AFI 1/2 over
> IPv4 BGP session, but seems next hop are not well populated (almost ::
> value)
>
> I'm wondering if this a bug (git version for fwdstatus patch) or if it's
> related to the fact we have AFI2/IPv6 over IPv4 BGP sessions.
>
> If I can select the BGP listening IPv4 / IPv6 address, I will be
> available to find if it's an issue or a PEBKAC ;)
>
> Thanks for this great tool ;)
>
> Le 2016-12-11 22:51, Fabien VINCENT a écrit :
>
> Hi Paolo,
>
> Happy to give you more work ;) It's not "needed", but it could be great I can
> do it.
>
> I thought about AF in another one, will try to implement / configure it on
> ASR and give the trick here for the community/list ;)
>
> Thanks for your help, great work and quick replies ;)
>
> Le 2016-12-09 17:54, Paolo Lucente a écrit : Hi Fabien,
>
> Yes, that is more clear, thanks for elaborating. In short, that is not
> currently possible. But the dev of the feature does not appear to be a
> biggie either, i've put it on my todo list. The only workaround that
> comes to mind, i know i'm saying the obvious, is to pick a transport
> (either v4 or v6, that is, not both) for BGP and travel ipv4/ipv6 AFs
> (along with any other AF you may need) in that same transport.
>
> Cheers,
> Paolo
>
> On Wed, Dec 07, 2016 at 09:53:54PM +0100, Fabien VINCENT wrote: Hi Paolo,
>
> Le 2016-12-07 17:54, Paolo Lucente a écrit : Hi Fabien,
>
> One step back on your question: you refer to the flow records or to the
> transport protocol here? You can travel v4 and v6 records within the
> same, say, v4 NetFlow/IPFIX/sFlow transport. This is what all exporters
> basically do; are you working with an exporter that is behaving in a
> different fashion? Like sending v4 records over a v4 transport and v6
> records into a v6 transport? If i understand your question correctly,
> that would be the only case in which it would apply.
>
> It can be, but this is not what I'm trying to achieve here.
>
> In any case you can select which v4 or v6 address you want to bind to,
> one (or otherwise you bind on all and then filter with iptables for
> example), with nfacctd_ip, sfacctd_ip and bgp_daemon_ip (the
> address you
> listen for NetFlow/IPFIX/sFlow can be different from the BGP one, yes).
>
> How can I see the BGP daemon to listen on a specific IPv4 and IPv6 ?
> I wanted to do it to have both address family.
>
> I run both nfacctd and sfacctd on the same server and BGP daemon
> have to be bind on sepecific seconday ipv4, because they have to
> bind both on TCP/179. And I want to do the same with IPv6 at the
> same time to have IPv6 sessions with DFZ routers.
>
> If this answer does not help, please elaborate a bit more on what you
> are trying to achieve.
> Is this more clear ?
>
> Thanks !
>
> Cheers,
> Paolo
>
> On Tue, Dec 06, 2016 at 01:21:22PM +0100, Fabien VINCENT wrote: Dear list,
>
> Any help on how to configure sfacctd and nfacctd on dual stack ?
>
> I mean having both IPv4 and IPv6 running on sfacctd, with collector
> enabled on both protocols ?
>
> I've seen the trick for bgp_daemon_ip, but don't know how to bind to
> specific ip4 and ip6 on my server the sfacctd bgp daemon and the
> daemon itself.
>
> Thanks in advance !
>
> ___
> pmacct-discussion mailing list
> http://www.pmacct.net/#mailinglists
> ___
> pmacct-discussion mailing list
> http://www.pmacct.net/#mailinglists
> ___
> pmacct-discussion mailing list
> http://www.pmacct.net/#mailinglists
Thanks, will have a look soon.
--
FABIEN VINCENT
---
@beufanet
---___
pmacct-discussion mailing list
http://www.pmacct.net/#mailinglists
