[rrd-users] Re: Errors in CGI's?: comparing 14all.cgi, routers2.cgi, mrtg-rrd.cgi

[Alex van den Bogaerdt]

On Wed, Nov 26, 2003 at 03:32:04PM -0800, Steve Smith wrote:
> Hi,
>     I am using mrtg 2-10.5, rrdtool 1.0.45, and have been testing the above
> three CGI's to verify their accuracy. All three use the same .rrd files. The
> CGI's are also the latest versions of each.
> I've notice that 14all.cgi and routers.cgi tend to agree with each other
> most of the time, but mrtg-rrd.cgi. is always lower with respect to the
> maximums, and differs with respect to average and current values.  The
> differences are not a consistent percentage. This occurs with several
> devices, an application, a switch, and a mail appliance. 14all provides
> logging for the graph commands which helps, but I haven't had time to
> capture the output other CGIs' graph commands so I'm not sure if they are
> plotting the same ranges or not.

Just some input in case someone else starts debugging:

    DEF:ds0:.....:AVERAGE
    ...
    GPRINT ds0:AVERAGE:"%6.2lf"
    GPRINT ds0:MAX:"%6.2lf"

produces different numbers than

    DEF:ds0:.....:AVERAGE
    DEF:ds0m:.....:MAX
    ...
    GPRINT ds0:AVERAGE:"%6.2lf"
    GPRINT ds0m:MAX:"%6.2lf"

unless the graph happens to be exactly as wide as the number of
rows fetched from the database.

This is a direct result from

MAX(AVG(a,b...),AVG(c,d...))
vs.
MAX(MAX(a,b...),MAX(c,d...))

cheers,
Alex
-- 
http://www.googlism.com/index.htm?ism=alex+van+den+bogaerdt&type=1

--
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