[rrd-users] Excluding UNKN or NAN from CDEF

[Eric Brander]

List,

I'm using the following graph building syntax that I borrowed from 
another member of this list. It takes the result of a web page parse and 
graphs the 1 or 0, and also adds a line for the service level % uptime.

rrdtool graph \www\graphs\webcheck\TEST_Uptimeday.png
--title="Uptime Last day, 5 Min Poll"
-w 700 -a PNG --start=-1d -e -600s -r -l 0.5
DEF:Response=\mrtg\log\WebCheck\TEST.rrd:ds0:AVERAGE
CDEF:Cleaned=Response,UN,0.0,Response,IF
CDEF:ResponseSum=PREV,UN,0,PREV,IF,Cleaned,+ CDEF:One=Cleaned,-1,GT
CDEF:OneSum=PREV,UN,0,PREV,IF,One,+
CDEF:Uptime=ResponseSum,OneSum,/
CDEF:UptDisp=Uptime,100,*
CDEF:AboveSLA=Uptime,0.9,GE,Uptime,UNKN,IF
AREA:Response#3399ff:"Response from Webserver"
HRULE:0.9#000000:"SLA 90 %\n"
LINE3:Uptime#FF0000:"UpTime below SLA"
LINE3:AboveSLA#00FF00:"Uptime Above SLA\n"
GPRINT:UptDisp:MAX:"Uptime Max %.2lf %%"
GPRINT:UptDisp:MIN:"Min %.2lf %%"
GPRINT:UptDisp:AVERAGE:"Mean %.2lf %%"
GPRINT:UptDisp:LAST:"Last %.2lf %%\n"
COMMENT:"Last Update: 09/27/2004 7:36:45 AM\r"

It works very well, except that since this is a new setup, the 
historical graphs have little data and the SLA line takes the NANs as 
failures. Again, I've taken this from someone else's example and I must 
admit that I really don't understand exactly what those statements mean.

My question to you: Is there a way I can build the SLA line without it 
incorporating NANs or Unknowns in to its equation?

aTdHvAaNnKcSe,

Eric Brander

--
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