[rrd-users] rrdfetch resolution problem

Matt Pounsett matt at conundrum.com
Thu May 3 03:33:40 CEST 2007

Hash: SHA1

On 2007-May-02, at 20:10, Alex van den Bogaerdt wrote:

> On Wed, May 02, 2007 at 05:02:22PM -0400, Matt Pounsett wrote:
>> Thanks for the clarification.. I understand what's going on now, and
>> should be able to work around it.  I think we'll wind up re-creating
>> these in order to get more of the data we want, later.
> Don't forget to read that fine manual on "rrdtool resize" !
> http://oss.oetiker.ch/rrdtool/doc/rrdresize.en.html

Oooh.. I hadn't spotted that.  Thanks, it will be extremely helpful.

> Do not increase the amount of rows significantly in one go.  Do this
> bit by bit, or else you will see NaN a lot when you query data that
> is available in other RRAs, but rrdtool selects the 5-minute RRA as
> a best match. Think about this for a while if you don't get it at
> first.

I think this would happen anyway, wouldn't it?  According to the  
docs, new rows are always added as old data... so if I were to double  
the number of rows, the RRA would just go back twice as far in the  
past, and the new stuff should be untouched.  Shouldn't it?

Version: GnuPG v1.4.6 (Darwin)


More information about the rrd-users mailing list