[rrd-users] Re: Graphing Very Small Numbers

Alex van den Bogaerdt alex at slot.hollandcasino.nl
Fri Jan 25 00:49:18 MET 2002

Cliff wrote:

> But I have to multiply by at least 100K in my mrtg configs.
> I assume this is because the 5-minute average that mrtg does
> by default produces an amount that is less than 1, which MRTG
> ignores because it's not a whole number.

Almost right.  MRTG doesn't ignore small numbers, it truncates
numbers to integers.  0.00001 isn't ignored, it is truncated to 0.

> So will RRDTool do a better job of graphing small numbers?

It will.
<checking source>
It can do atobytes if you wish (that's 10^-18).

> Or how are other folks graphing numbers that 5-minute-average out to be
> less than 1?

I have graphed such a counter as a percentage of a declared maximum
allowed number.  If you think 10 errors per 300 seconds is the limit,
10 should be equal to 100% and thus each error should be 10%  This
comes down to multiplying the counter with 3000.  If you get one
error in an interval, mrtg will calculate 3000/300 == 10.
It is probably best to use unscaled[] for these kind of images.

 / 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:rrd-users-request at list.ee.ethz.ch?subject=unsubscribe
Help        mailto:rrd-users-request at list.ee.ethz.ch?subject=help
Archive     http://www.ee.ethz.ch/~slist/rrd-users
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi

More information about the rrd-users mailing list