[Perfparse-users] Re: Perfparse graph scale problem

[Hans E. <eng...@gm...>]

I had a similar problem. I preferred to change the value reported by
the plugin instead. It's a little more work maybe but in the end made
more sense too. Instead of having both Nagios and Perfparse report
memory in bytes for example I now routinely convert these values to
kilobytes and megabytes instead. This makes the values more readable
all round. Of course this might not be possible for all values so. The
same could be done for your T1 line unless of course you need such
high resolution (i.e. bytes/sec) on your measured values.

Either way it's an alternative you could consider.

Cheers,
Hans

On 6/29/05, Ben Clewett <bcl...@pe...> wrote:
> Ben,
>=20
> The code for this may be found in the function 'drawYScale()'.  Look for=
=20
> line 1031 in file cgi/perfchart.c.
>=20
> Ben.
>=20
>=20
> Ben Conrad wrote:
> > Hi Folks,
> >=20
> > I'm using Nagios + Perfparse to graph bytes/sec on some of my networkin=
g
> > equipment (T1 speed).  It seems that when the data value goes over a
> certain
> > number (100,000??) that the Perfparse graphs switch to using exponents =
on
> > the Y-axis.  So right now for my T1 usage of about 138785 (bytes/sec)
> > Perfparse displays this value as 1.80+e05 on the chart, it doesn't read
> > very well.
> >=20
> > Is there any way to change this so that the actual value stored in the =
Db
> is
> > displayed in the graph?
> >=20
> >=20
> > Thanks,
> >=20
> > Ben