Just hit this one.

There is no special rrdtool create/graph that triggers this because it's not 
rrdfile specific issue.

Basically im->magfact is not set if you select --logarithmic

The only code that sets  im->magfact is si_unit()

si_unit() is not called if --logarithmic

so the value in the structure is indeterminate for log graphs.

then in calc_horizontal_grid()   scaledrange is calculated based on the the 
range and the im->magfact

so if im->magfact is NaN  or some other value (0?) that makes scaledrange NaN 
then the isnan test
immediately following will fail and no graph is produced.

a quick hack in auto_scale() seems to ensure that graphs are produced. 

   im->magfact = (*magfact);

Whilst the graphs don't look too ridiculous I've not had a chance to do any 
verification. I'm also not sure
this solution is strictly correct given the multiple calls to auto_scale() for 
the second Y annotations.

-- 
Bernard Quatermass.
_______________________________________________
rrd-users mailing list
[email protected]
https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users

Reply via email to