[mrtg] Re: wrap redux
alex at ergens.op.Het.Net
alex at ergens.op.Het.Net
Mon May 29 11:15:45 MEST 2000
Randy Bush wrote:
> i believe i am getting 32-bit counter wraps on oc12s and above. the list
> archive does have a thread on wraps, but it seems inconclusive or i am just
> not getting it. anyone have some clues for me?
What is it that you don't understand?
Just calculate the maximum value. For a 32-bit counter this is 2**32-1
which is 4,294,967,295 octets. If another octet is transfered, the
counter wraps to zero and starts counting again.
If a counter wrap occurs, two things can happen:
1) the new value is equal or higher than the previous value. MRTG
will not know that the actual change is 4MB higher than it calculated.
2) the new value will be lower than the previous value. MRTG will
consider this to be a faulty value and the previous rate is used.
Suppose that the first case will happen followed by the second case.
If you measure every 300 seconds (5 minutes) and the data rate is at
2**32+1 per 300 seconds, it will look as if the counter incremented
with only 1. Hereafter the data rate changes to 2*32-1 which means
that every time the counter is read it looks as if the counter is
running backwards by 1 each time. MRTG detects this as a faulty value
and keeps the 1.
Please ask a more specific question if you need more help.
regards,
--
__________________________________________________________________
/ 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 |
+----------------------------------------------------------------------+
| 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
More information about the mrtg
mailing list