[132105] in North American Network Operators' Group
Re: flow analysis for juniper devices
daemon@ATHENA.MIT.EDU (Mehmet Akcin)
Sun Nov 14 17:24:18 2010
From: Mehmet Akcin <mehmet@akcin.net>
In-Reply-To: <20101114151223.GA3060@moussaka.pmacct.net>
Date: Mon, 15 Nov 2010 08:23:42 +1000
To: Paolo Lucente <pl+list@pmacct.net>
Cc: nanog@nanog.org
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
Hi again,
I think many of you got stuck at the "freeware" part however if there =
are good suggestions you have which would help achieving what I would =
like to achieve by paying for a software, I am open to listen =
suggestions on that as well.
just thought would mention
mehmet
On Nov 15, 2010, at 1:12 AM, Paolo Lucente wrote:
> On Sun, Nov 14, 2010 at 09:04:36AM -0500, Jared Mauch wrote:
>>=20
>> On Nov 14, 2010, at 3:59 AM, Paolo Lucente wrote:
>>=20
>>> OTOH it would be nice to see one day those NetFlow v9 MAC address =
fields
>>> populated on higher-grade boxes, say, to facilitate analysis of =
public
>>> peering at internet exchanges ...
>>=20
>> This can be done with mac accounting in your platform. It may =
require ordering the right PIC in this case, but is easy to collect that =
data.
>=20
> MAC accounting is not a solution, except for basic things, for at
> least two reasons: 1) it is coarse-grained and 2) its segregation
> from flow accounting prevents having all the eggs in one basket.
>=20
> Cheers,
> Paolo
>=20
