[rrd-users] Stacking at Infinity

Martin Sperl rrdtool at martin.sperl.org
Sat Mar 3 07:26:02 CET 2012

I have not fully understood how you want your graph to look visually.
But maybe you can use TICK to draw your errors.

I remember that I have done so to highlight times when we have exceeded predictions.
You could obviously paint multiple TICKS for different types and paint them with different colors using transparency.
As the color (de)coding may be a pain you possibly could draw multiple ticks of different fractions.


On 03.03.2012, at 03:46, Alex van den Bogaerdt wrote:

>> LINE3:DSxH#xxxxxx
>> This will draw a line at INF using the classes colour if errors were
>> detected in that class. If multiple classes have drops then the line
>> will get overwritten with the later colour. I could use transparency
>> but then it is usually difficult to determine the colours. I have
>> tried stacking, but it doesn't like stacking on top of infinity which
>> sort of makes sense.
> By definition you cannot graph *at* infinity. That is why this "value" was
> chosen to trick RRDtool to forget about autoscaling and just graph the
> portion already visible.
> Not only is stacking "on top of" infinity not defined, you would also, by
> definition, stack above (or under) the visible part of the graph and thus
> outside the viewport.
> What you want is the top of the graph, and infinity is just the tool you
> use to get there.
> Unless somebody extended the rudimentary VDEF framework, there currently
> is no way to find out dnamically what autoscaling is doing. But if you
> find out by other means (e.g. a double pass run, or by having a fixed
> maximum (as you seem to have)) then you can use that information to draw
> your lines at e.g. 95% of maximum and 90% of maximum. It's not exactly
> what you aim for now, but I hope it is matching what you're really after.
>> If I had drops in two classes then I would like to see 2 lines stacked
>> on top of each other in different colours at infinity or negative
>> infinity for inbound drops.
> One line at the top (or bottom) of the graph, then stack a suffient amount
> in the "wrong" direction. I never tried it, but stacking -50M "on top" of
> 150M should get you at 100M.
> Alex
> _______________________________________________
> rrd-users mailing list
> rrd-users at lists.oetiker.ch
> https://lists.oetiker.ch/cgi-bin/listinfo/rrd-users

More information about the rrd-users mailing list