[rrd-users] RRDs::graph: lost points on resulting graph

bitman bitman at webazilla.com
Wed May 9 11:06:46 CEST 2012


Hello, S Shipway

I'm completely certain that the versions of the executable and the Perl
library are not the same: 1.4.5 != 1.4004.





Both the executable and the library were installed from FreeBSD ports to the
package rrdtool-1.4.5. I thought that the sequence of versions would be
correct even inspire of different version numbers. Isn't in right?

The structure of the RRD file (built by cacti) used to render both graphs:



> The graph values look substantially different, though the shapes are
> similar.
> Possibly you have different multipliers or you are using different
> consolidation functions.

That's the problem. How comes the difference in multipliers or consolidation
functions while using the same RRD file and the same parameters? The only
difference is the grapher interface, which should have used same core libs.

Is there a way to explicitly control those critical parameters at render
time?

The actual code of the Perl part which gives wrong graph:



--
View this message in context: http://rrd-mailinglists.937164.n2.nabble.com/RRDs-graph-lost-points-on-resulting-graph-tp7538827p7542295.html
Sent from the RRDtool Users Mailinglist mailing list archive at Nabble.com.



More information about the rrd-users mailing list