[mrtg] Re: Kb/s or b/s

[Alex van den Bogaerdt]

Kyrre Baker wrote:
> 
> > Great.  What did you change?
> 
> I had to figure out the "target" I had to gather the data from. Luckely
> I allready have some knowledge for the 801, and snmpwalked both the
> routers and found out what the differences and right targets where.
> Here`s the target:
> 
> Target[Cisco_1]:
> 1.3.6.1.4.1.9.10.1.1.1.2.1.1.1.0&1.3.6.1.4.1.9.10.1.1.1.2.1.1.2.0:public
> @10.0.0.1

This is the experimental part of the cisco specific MIB. Beware that
this will most likely change.  As my mib browser does not know about
this specific part of the tree, I cannot comment on it any further.

> > What I notice at a first glance is the huge (compared to the 
> > rest) spike at 19:30.  This causes the rest of the traffic to 
> > become virtually invisible.
> 
> Yes. That's right. The reason is (I think) that this network is resident
> at our home, and noone was awake in the "invisible" timearea. :)

Hmm...  Inefficient network usage.  Shame on you :)

> > Try to figure out if
> > - there is no traffic that you didn't take into account
> 
> Hm. Don't thing so. DHCP, Sendmail, POP, HTTP and the usual ports are
> alle routed via the Cisco.

And what about the traffic from the router/to the router?  For instance,
how much data is collected doing a snmpwalk ?

> > What happens if you start a huge download.  Do the two graphs 
> > show the same after a while?
> 
> No. What I cant seem to find out is if it`s k/s *or* kb/s that's the
> right numbers for the FreeBSD og for the Cisco. When the traffic
> increases on the Cisco, it also increases on the FreeBSD (and vicevera).
> There`s only the matter off which of the graphs shows the "right"
> prefix.

Perhaps you're monitoring packets, not octets, on the cisco.  If so,
compare the number multiplied by about 800 (Assuming 800 is the
average packet size).

You could try to run ping for a while.  Let it run for 15 minutes
with a packet size of 1000.  This should give you at least one complete
interval with a rate of about 1000 per second.  Don't access the web
by other means, and compare the graphs again.
Repeat with a packetsize of 500.  You should get 500 Bps.  Now, if I'm
right about the packets vs. bytes problem, you should see no difference
on the cisco and you should see the traffic decrease on the BSD box.

cheers,
-- 
   __________________________________________________________________
 / alex at slot.hollandcasino.nl                  alex at ergens.op.het.net \
| work                                                         private |
| My employer is capable of speaking therefore I speak only for myself |
+----------------------------------------------------------------------+
| Technical questions sent directly to me will be nuked. Use the list. | 
+----------------------------------------------------------------------+
| http://faq.mrtg.org/                                                 |
| http://rrdtool.eu.org  --> tutorial                                  |
+----------------------------------------------------------------------+

--
Unsubscribe mailto:mrtg-request at list.ee.ethz.ch?subject=unsubscribe
Archive     http://www.ee.ethz.ch/~slist/mrtg
FAQ         http://faq.mrtg.org    Homepage     http://www.mrtg.org
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi