Usually I can deal with these, but this one is odd…
Our perfectly operating nagios installation simply stopped working – no changes or upgrades, just quit. It now gives this error:
Error in configuration file ‘/web/tools/nagios/etc/nagios.cfg’ - Line 18 (NULL value)
Although I haven’t attached the nagios.cfg file, I can… HOWEVER, what is happening is that it will give that error on the very first directive it hits in the cfg file no matter what it is or where it is. Even creating a test cfg file with only one definition, no comments, etc., it will give the above error on then line 1.
Nagios version 2.0b3 on Linux SLES 9. It’s even been completely reinstalled and new configurations built with the same results.
Any clue on what’s causing this behavior? we’re stumped.
Edited ]
I have to assume that this is the output you get when running the -v option to nagios. I also have to assume you meant that you recompiled nagios from source. Linux SLES 9. is what flavor?
Anyway, it sounds like someone has removed software/libs. Double check the output from your fresh compile for errors.
You ran make clean first right?
Oh, absolutely. We even pulled a completely fresh release, just to make sure we also had the latest version.
The configure, make and install on all components (application, plugins, etc.) are all very clean. No complaints (save for optional graphic libraries that weren’t there before anyways).
It’s also odd that the behavior started with an installation that was working but without change, stopped. It leads me to believe that something more system level is involved, but for the life of me, I can’t figure anything that would impact nagios in this way and just nagios…
BTW: SLES 9 is SuSE Linux Enterprise Server, version 9. (2.6.5-7.151-smp for those who prefer that nomeclature :(.
I’m new to selinux, but for now, if you have it, can you turn it off and see if it helps you out?
I’m not certain what you want me to turn off?