Nagios 2.12 locks

[awd]

Hi there,

We have installed nagios on 2 servers to monitor our network, however since the last portupgrade, nagios on both machines just sits there, and doesn't probe any machines or services.


The logs don't say anything other than
[1213758202] Nagios 2.12 starting... (PID=77539)
[1213758202] LOG VERSION: 2.0
[1213758202] Finished daemonizing... (New PID=77540)
[1213758218] Caught SIGTERM, shutting down...

The sigterm doesn't actually kill it, it still just sits there. I can't kill the process even with a -9.

Anyone seen this or have any ideas?

# uname -a
FreeBSD my.webzone.net.au 5.4-RELEASE FreeBSD 5.4-RELEASE #0: Mon Oct 30 22:18:18 CST 2006 root@my.webzone.net.au:/usr/src/sys/i386/compile/KERN i386

# pkg_info nagios*
Information for nagios-2.12_1:

I am aware of the EOL on 5.x however both servers nagios' stopped before the EOL came into play.

TIA
Cheers
cya
Andrew

[KernelPanic]

I currently have Nagios 2.12 running on FreeBSD 6.3 i386 and FreeBSD 7.0 amd64 without any problems.

When FreeBSD 7.0 first came out I tried running Nagios on an i386 system. While the main process did not go crazy I started having child processes hang just like that (i.e. unkillable zombie) Rather than mess with it I rolled the box back to FreeBSD 6.3.

Nagios' use of fork and FreeBSD have not had a very cordial relationship.

You may want to try and hit the nagios-users mailing list and/or email the maintainer to see if he might have suggestions.

[awd]

Originally Posted by KernelPanic

I currently have Nagios 2.12 running on FreeBSD 6.3 i386 and FreeBSD 7.0 amd64 without any problems.

I've got one to work, a warning about a contact not associated a group or service. Sorted that out and away it went. Beforehand it didn't worry about it
I have nagios 2.12 on a number of other sites without issue.

Originally Posted by KernelPanic

You may want to try and hit the nagios-users mailing list and/or email the maintainer to see if he might have suggestions.

ok thanks

[scottro]

Assuming you're more or less familiar with nagios, you probably realize that warning isn't that important, it can just refer to a contact that you've put in and aren't using yet. Warnings are fine, errors are bad.

KP, good to see you. You're fortunate that I didn't know you knew nagios several months ago.

Although it's for a different O/S I found Max Hetrick's CentOS wiki howto very good for coming to grips with nagios. (In my case, I'm running it on CentOS anyway, but Max breaks down what each file does in a way that I found very easy to follow.)
Too lazy to give the link right now, but it's somewhere on wiki.centos.org.

[awd]

Originally Posted by scottro

Assuming you're more or less familiar with nagios, you probably realize that warning isn't that important, it can just refer to a contact that you've put in and aren't using yet. Warnings are fine, errors are bad.

I've been using it for about 3 years and I can pretty much monitor anything , provided perl is installed. I realise that warnings are 'fine', however when nagios starts working after associating a contact with a group when it wasn't b4 (which is just a warning), indicates something is a miss.

I'd be interested to see if someone could put a contact into their nagios and not associating it with a machine or group, to see what happens.

Cheers
cya
Andrew

[scottro]

Probably just a warning, I assume. I only have a production machine with it right now, so can't really experiment.