[mrtg] Re: ISP cheating...

[Jay Hennigan]

On Mon, 19 Jun 2000, Fernando Medina wrote:

> Our ISP uses MRTG to monitor our traffic on our link.  We also use MRTG
> so we monitor out local serial port and they monitor their own serial
> port.  We have access to their graphs and we are seeing a big
> discrepancy in traffic.  After trying to figure it out, I notice that
> they are updating their graph every 2 minutes, but they still say 5
> minute average on the page.  If tehy simply changed their cron job to
> run every 2 mintues instead of every 5 and did not change the actual
> MRTG program itself it would show more traffic than there actually is,
> wouldn't it?  We are on a test basis with thsi provider and I get the
> feeling he is trying to screw us... thanks,

A few possibilities...

The more frequently the samples are taken, the higher the peak graph will
be on bursty traffic.

For example, assume you download a large file that takes three minutes, 
and you use 100% of your port speed for that time.  During the next two
minutes you are idle. 

With samples taken every two minutes, you'll show a spike to 100% of the 
port during the download.  With samples taken every five minutes, the 
graph will show 60%.  The graph with the faster sampling rate will tend 
to be more jagged, with higher peaks.  

However, the average numbers should be the same or at least relatively 
close at the end of the day/week/month.  

The other considerations are in vs. out interfaces which will obviously
be reversed at both ends of the link, and bits vs. bytes.

-- 
Jay Hennigan  -  Network Administration  -  jay at west.net 
NetLojix Communications, Inc.  NASDAQ: NETX  -  http://www.netlojix.com/
WestNet:  Connecting you to the planet.  805 884-6323 


--
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