[rrd-users] Re: Let's say I'm collecting 5 min (300 sec) data points. My next database is the 30 min (1800 sec) average. Assuming good data for 30 minutes, I see that Cacti sums up the data points and divides that by 6 points. What if just 1 of the data points has N

[Alex van den Bogaerdt]

On Fri, Nov 04, 2005 at 02:56:10PM -0600, Scott Moseman wrote:
> Let's say I'm collecting 5 min (300 sec) data points.
> My next database is the 30 min (1800 sec) average.
> 
> Assuming good data for 30 minutes, I see that Cacti
> sums up the data points and divides that by 6 points.

Cacti doesn't, RRDtool does.  Actually I *am* guessing here but it
is an educated guess.

> What if just 1 of the data points has NaN for its value.
> What's the logic now? Sum the known values and use
> 5 as the divisor? Assume its 0 and still divide using 6?

If you let it, RRDtool will take the average of those five.  See
the manual about xff.

If you want unknowns to be treated as zero, you need to do the
CDEF trick as shown in my tutorial.

HTH
-- 
Alex van den Bogaerdt
http://www.vandenbogaerdt.nl/rrdtool/

--
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://lists.ee.ethz.ch/rrd-users
WebAdmin    http://lists.ee.ethz.ch/lsg2.cgi