nanog mailing list archives

FW: Graphing Peering


From: Daniel Golding <dgolding () burtongroup com>
Date: Fri, 21 Jan 2005 15:01:16 -0500



Additional information on MAC accounting from Hakan Lindholm...

(specifically, the SNMPv2c object to pull 64bit MAC accounting counters)

- Dan

------ Forwarded Message
From: Hakan Lindholm <hakan () staff spray se>
Date: Fri, 21 Jan 2005 20:36:45 +0100 (CET)
To: Daniel Golding <dgolding () burtongroup com>
Cc: <druid () softdust com>, andrew matthews <exstatica () gmail com>
Subject: Re: Graphing Peering

I'm not registerred to post on nanog.
You may send this info in, with or without quoting me..

On Thu, 20 Jan 2005, Daniel Golding wrote:


Andrew,

The 32 bit counters are a significant problem when using gigabit ethernet
public peering interfaces. Needless to say, MAC accounting was not designed
for gigabit speeds. Frequent polling is, sadly the only solution. If you
write your own scripts, make sure to account for counter wrapping.

What about the .1.3.6.1.4.1.9.9.84.1.2.3.1.2 tree?
Remeber to use SNMPv2c.

We use the following to generate some MRTG config:


while (!$session->{ErrorStr} and
        $$vars[0]->tag eq "ipNetToMediaNetAddress"){

     if ($type eq "dynamic") {

         @mac = split(/:/, $mac);
         $decmac = join('.', hex $mac[0], hex $mac[1], hex $mac[2], hex
$mac[3], hex $mac[4], hex $mac[5]);
         ($iname, @junk) = gethostbyaddr( pack( "C4", split( "\\.", $ip )),
AF_INET );

         if (-z $iname) {$iname = $ip};
         if (!defined($peers{$ip})) {$peers{$ip} = "no BGP peer"};

         $ifi = $ix{$router}[1];

         print "\n";
         print "Target\[$ip\]:
1.3.6.1.4.1.9.9.84.1.2.3.1.2.$ifi.1.$decmac\&1.3.6.1.4.1.9.9.84.1.2.3.1.2.$i
fi.2.$decmac:$ARGV[1]\@$ARGV[0]:::::2\n",

         "MaxBytes\[$ip\]: 25000000\n",
         "Title\[$ip\]: $ix{$router}[0]: $peers{$ip}\n",
         "PageTop\[$ip\]: <H1>$ix{$router}[0]: $peers{$ip}</H1>\n",
         "\tIP: $ip, DNS: ", $iname, "\n";
     }
     ($ip,$mac,$type) = $session->getnext($vars);
};

(This is only part of the script.  You should make it work in your
environment quite easy though.)


- Dan

on 1/20/05 9:45 AM, "druid () softdust com" <druid () softdust com> wrote:


On Wed, 2005-01-19 at 22:41, andrew matthews wrote:

Another problem you might run into is counter wrapping. When polling
every 5 minutes, some counters may wrap. (there is no 64 bit counter for
the mac-address accounting). So you have to run it in short timeframes,
causing more cpu utilization.

Talking about Cisco, see above.  There is such counters.


But all in all, mac-accounting and Netflow source-as give you a very
good overview of your network flows.

Yes indeed.

/H

------ End of Forwarded Message


Current thread: