On Fri, May 09, 2008 at 07:25:45PM +0200, Tobias Oetiker wrote: > Hi Bigunyak, > > the 'problem' is that the on-disk data format of rrdtool does not > lend it sels to adding and removing data sources ... all that can > be done sensibly is to reqrite the file in its new structure ...
Which will be faster than the dump-edit-restore procedure. But who really needs that? > another thing people often ask is to changeing the step size of an > archive ... it would be an interesting task to write a rrdtool > re-sampling mechanism which is able to alter the step size of an > existing archive while keeping the data as intact as possible ... Averages won't be a problem, but how to deal with minimum,maximum and last? One thing's for sure: without the original input, any change will be guesswork except for the special case where the new step size is a whole integer times the old step size. E.g. LAST: Consider 300 seconds per interval, being changed into 200 seconds per interval. Each time the rate was 0 for the larger part of the interval, except at the end (0,0...0,1,1},{0,0...0,1,1} -> 1,1. Result: 600 seconds rate 1, in two intervals. The sensible thing to do would to change this into 3 intervals of 200 seconds, also with rate 1: 1,1 -> 1,1,1. But, with the original input available, the sensible thing to do would be to change this into two intervals with rate 0 and one with rate 1: {0,..,0},{0,..,1,1,0,0...,0},{0,..,0,1,1} -> 0,0,1 E.g. MAX: The same input as above would result in 1,1 being changed into 1,1,1 but with the original input available the result would be 0,1,1. (0,0...0,1,1},{0,0...0,1,1} -> 1,1 -> 1,1,1. {0,..,0},{0,..,1,1,0,0...,0},{0,..,0,1,1} -> 0,1,1 I think you would be opening a can of worms. Alex _______________________________________________ rrd-developers mailing list rrd-developers@lists.oetiker.ch https://lists.oetiker.ch/cgi-bin/listinfo/rrd-developers