--- Cooper Nelson <[EMAIL PROTECTED]> wrote: > I'm experimenting with the Holt-Winters abberant behavior detection in > the current dev. release. I'm curious about graphing the number of > confidence band violations per hour rather than just the binary failure > marker. I can set the failure threshold to its most sensitive setting > (1 confidence band violation per failure) to create a RRD of binary > confidence band violations; but I cant figure out how to graph them as a > sum over an hourly period (0-12 per hour assuming 5 minute averages). > > Would creating a new RRD with hour averages and populating it with the > failures work? Is there a better way?
Hi Cooper, I think creating a separate RRD would be a simplier way than incorporating a new functionality into rrdtool. Perhaps, this copule of my scripts (at the bottom of the page) would be of help to crteate your own: http://carrierin.sourceforge.net I've even created a new SourceForge project for such experiments, but never had a time to play with that: http://sourceforge.net/projects/behan/ If you find any of your scripts worth publishing, I can put them there too. Cheers, Stan __________________________________________________ Do You Yahoo!? Yahoo! Finance - Get real-time stock quotes http://finance.yahoo.com -- Unsubscribe mailto:[EMAIL PROTECTED] Help mailto:[EMAIL PROTECTED] Archive http://www.ee.ethz.ch/~slist/rrd-developers WebAdmin http://www.ee.ethz.ch/~slist/lsg2.cgi
