From: Yves M. <yme...@li...> - 2004-07-30 13:14:14
|
> BTW, this may also be relevant: I use a different approach for writing > the serviceperf.log file, because I used the default (command based) > method before I started using perfparse, and stuck with that. > I choose to go that route because I allready had written some custom > tools that I used to do some reporting, based on the performance data > output files, and would lioke to keep that as well, so I ended up using > a command that will create both files for me. > This also means the nagios process is not actually locking the file, > so I'm now considering the following: > Rename the file to a temp file. A new file will be started > automatically, no kill signals required. > Read in the temp file, and delete it. perfparse already does this, except that it also sends the HUP signal. This is necessary because without this, with the temp file renamed (no in= ode change), nagios would continue to write to the temp file instead of recreating a n= ew file. Yves --=20 - Homepage - http://ymettier.free.fr - http://www.logicacmg.com - - GPG key - http://ymettier.free.fr/gpg.txt - - Maitretarot - http://www.nongnu.org/maitretarot/ - - GTKtalog - http://www.nongnu.org/gtktalog/ - |