[rrd-users] results with 'm' and 'u' and converting 'nan' to '0'

[ivan jr sy]

Hi,

Im trying to create rrd to graph DNS queries per second. I need help, to look for the reason why the graph shows "m" and "u" values, what are those? and how to eliminate them, the figures too are incorrect.

overview:
every 1 minute, the cron will use rndc stats to show the DNS success and failure counters. Unfortunately there are times where the DNS engine (BIND) gets restarted and rdnc stats counter resets to zero..

After reading through "NOTE on COUNTER vs DERIVE" on http://oss.oetiker.ch/rrdtool/doc/rrdcreate.en.html

I believe i got it *partially*, to use "DERIVE" with min=0. So here's what ive done so far.

/usr/local/bin/rrdtool create dns.rrd -s 60 \
         DS:rndc_success:DERIVE:80:0:U \
         DS:rndc_failure:DERIVE:80:0:U \
         RRA:AVERAGE:0.5:1:120 \
         RRA:AVERAGE:0.5:5:288 \
         RRA:AVERAGE:0.5:30:336 \
         RRA:AVERAGE:0.5:120:360 \
         RRA:AVERAGE:0.5:720:730


The graph:
.
.
.
         DEF:Success=dns.rrd:rndc_success:AVERAGE \
         LINE1:Success#0000FF:" Successful" \
         GPRINT:Success:LAST:"current\: %2.0lf%s"  \
         GPRINT:Success:MAX:"max\: %2.0lfqp%s"  \
         GPRINT:Success:AVERAGE:"average\: %2.0lf%s\n"  \
         DEF:Failure=dns.rrd:rndc_failure:AVERAGE \
         LINE1:Failure#FF0000:" Failure" \
         GPRINT:Failure:LAST:"current\: %2.0lf%s"  \
         GPRINT:Failure:MAX:"max\: %2.0lfqp%s"  \
         GPRINT:Failure:AVERAGE:"average\: %2.0lf%s\n"  \

there are *some* figures which results to:

current: 2
maximum: 292m
average: 123m

(there even once when it showed)
average: 124u

- what are the "m" and the "u" for? is this the overflow? how do i get rid of those and compute correctly? im missing something a lot here...

Another thing (not related above)
there are times where graph from previous, when the value of GPRINT prints the string "nan" in the graph, how do i convert the string 'nan' to the string "0" ?

Thank you for your time.