hole in the argus archive on theorygroup.org

[Mark Poepping]

Are you going to ship more than 8-900 Mbps so you'll need bonding?  Or
are you just planning ahead?  Do you know of a system that can ingest
that?  I guess that's what you want to find out..  We haven't tried
choking our fastest stuff yet..

We're doing everything on linux (2.4.16 right now), mostly because we're
lazy and it's been keeping up (using Carter's commercial code -
gargoyle, which kicks butt on the argus stuff by the way).

We have several GigE cards, seem to have the best luck with the
SysKonnect fiber, but have both generation Intel (the second gen were
*much better*, but we've had some issues with them as eth2) and some
3Com's too.  Had the 2-port SysKonnect for a while, but it seemed to be
designed more for redundancy than performance, so I shipped it back.  I
admit we haven't been too systematic about the analysis mostly lacking
money to create time for that diligence. 

There seems to be a fair disparity in the interrupt load of the various
cards, which isn't surprising, but I haven't had the time to look into
real counts (from /proc/net since netstat can't count too high).

We'd talked about trying to do some more benchmarking on performance
numbers, but it wasn't much of a priority in our conversations with
Carter, especially since it's pretty easy to catch up by splitting
flows..

Mark.


> -----Original Message-----
> From: owner-argus-info at lists.andrew.cmu.edu [mailto:owner-argus-
> info at lists.andrew.cmu.edu] On Behalf Of Peter Van Epp
> Sent: Thursday, February 28, 2002 11:20 AM
> To: argus
> Subject: Re: hole in the argus archive on theorygroup.org
> 
> 	For at least me, it wasn't only the archive that missed these. I
> don't find any of the three posted so far in my mail file either (and
I'm
> quite interested because dark fibre and a GigE link are in my and more
> importantly in my argus's short term future. Did others on the list
see these
> or was there a black hole for a while?
> 	While I'm here a suggestion I came across elsewhere: the Linux
channel
> bonding code from George Becker works with GigE cards and can combine
two
> GigE outputs in to a single logical device for feeding to things (such
as
> for instance argus). In combination with the "Turbo" zero copy packet
capture
> code in Linux this may be interesting.
> 	Which operating system is the CMU argus running on (I don't see
it in
> the interesting list of machines I should be considering buying :-))?
> 
> Peter Van Epp / Operations and Technical Support
> Simon Fraser University, Burnaby, B.C. Canada
> 
> >
> >
> > There aren't any posts recorded in the archive between 2/8-2/13..
Was
> > there a problem in there?  I know there were a few posts, especially
> > some responses to Russell's "Giving a talk on Argus" query -
including
> > the one fairly extensive post detailing some of our configuration
work
> > at CMU..
> >
> > Carter, do you have a copy that perhaps can be reposted on the
archive?
> > Mark.
> >
> <snip of interesting post>