<snip> > The create syntax. > ./rrdtool create ../data/apollo.webmagix.net.cpugebruik.rrd -s 300 \ > DS:user:COUNTER:600:0:100 \ > DS:system:COUNTER:600:0:100 \ > DS:idle:COUNTER:600:0:100 \ > RRA:AVERAGE:0.5:1:288 \ > RRA:AVERAGE:0.5:6:336 \ > RRA:AVERAGE:0.5:24:360 \ > RRA:AVERAGE:0.5:288:365 > <snip> > The values inserted with the update are seconds (from /proc/stat). So I > expected that in the .rrd there would be values with a total of 300 (user + > system + idle = total uptime of the system; iow if I update it every 300 > secs I expect it to be a total of 300 every time) But when I fetch the .rrd > i get stuff like this: > > 1029347700: 1.3564594315e+00 5.9976522702e-02 5.8354189738e-01 > 1029348000: 1.3445628391e+00 5.6877146042e-02 6.0188219993e-01 > 1029348300: 1.3267260498e+00 5.0022853958e-02 6.2326224452e-01 > 1029348600: 1.3333111111e+00 7.3255555556e-02 5.9343333333e-01 > 1029348900: 1.3090513104e+00 4.9912218531e-02 6.3441410114e-01 > > Which isn't correct as far as I can see.
rrdtool automatically consolidates the data into "/s" (per second). That's because you use counter DS's. So, it should be exactly count up to 1 per second (as you have 1 sec CPU-time per sec time to share between user/sys/idle) But: Your DS's make: 1.3564594315 + 0.0599765227 + 0.58354189738 =~ 2.0 Ergo: You have a dual-cpu system. Regards, Geert van der Ploeg -- Unsubscribe mailto:[EMAIL PROTECTED] Help mailto:[EMAIL PROTECTED] Archive http://www.ee.ethz.ch/~slist/rrd-users WebAdmin http://www.ee.ethz.ch/~slist/lsg2.cgi