0

I wrote my own munin-plugin and I'm confused, how munin represents values between 0 and 1. plugin values The 2nd line of values uses an notation with 'n'. What does it mean and how can I avoid it? I just want a common floating point value like 0.33! What I'm doing wrong?

Here is the plugin configuration:

graph_title Title
graph_args --base 1000 -l 0
graph_vlabel label
graph_category backup

Every hint is welcome.

UPDATE

Ok, I finally found it: https://serverfault.com/questions/123761/what-does-the-m-unit-in-munin-mean

'm' stands for milli!

I am a bit confused, why munin is using it in this context!

Is there any possibility to avoid this?

Thilo Schwarz
  • 640
  • 5
  • 24

1 Answers1

1

Some configuration items that may be of interest. See Munin Plugin Reference.

You may change the resolution in numeric resolution in the table data with the below. Use C/C++ formatting, and spacing may be an issue in end result.

graph_printf %5.2le     # default is %6.2lf
                        # appears l (for long) is needed

For certain graph types you can also alter the display period for the munin graph. Munin graphs the average data between last two (nominally 5 minute spaced) measurements - with default of per/sec units on the result. This is based on the var.type attribute which defaults to GAUGE. If your for example data is reporting accumulated counts you can use:

type DERIVE         # data displayed is change from last 
                    #   reading 
type COUNTER        # accumulating counter with reset on rollover

for each of these you can change the vertical axis to use per/minute or per/hour with:

graph_period minute # if type DERIVE or COUNTER  
                    # display avg/period vs default of avg/sec

Note that the table shows the derived data for the graph_period.
See RRDCreate == rrdtool create for implementation particulars- munin used rrd underneath.

user1725445
  • 63
  • 1
  • 5