[collectd] collecting nagios perfdata

[Łukasz Kostka]

Hi

I want to collect perfdata from nagios. Nagios can either run a command 
for every perfdata output, or append values to a file. I think that it 
will be easier to append data to a file. For collecting information i 
can write a custom python plugin, or tail the file.

1 Is it possible to run a python3 plugin ? I could not find any 
information on how to force collectd to run python3 interpreter.
2 If i will tail a file i can use tail or tail_csv plugin. In both cases 
I need to keep perfdata output in separate files per check command. 
Collectd tails file for new values, but does it "empty" the file once 
values are read ? Do I need to "empty" the file from time to time to 
prevent continous file growth ?