Nagios on OpenBSD

[mksmr]

Hi @all, I am currently switching a number of machines from Linux to OpenBSD and got a bit stuck over the Nagios Monitor. OpenBSD6.0 comes with Nagios Core 4.0.8. I made a fresh install on a vm and installed Nagios per pkg_add, then followed the (aged) instructions at http://www.kernel-panic.it/openbsd/nagios/index.html which mainly applied to OpenBSD 4.8. Nagios itself is running now, and I can authenticate with user/passwd and (partially) access the initial page. Still, this page reports Nagios as "not running" and the side panel is pretty messed up. This is what the whole thing looks like (see attachment).

This is my apache conf module:

Code: Select all

ScriptAlias /nagios/cgi-bin "/var/www/cgi-bin/nagios"

<Directory "/var/www/cgi-bin/nagios">
    Options ExecCGI
    AuthName "Nagios Access"
    AuthType Basic
    AuthUserFile /var/www/etc/nagios/nagios.passwd
    Require valid-user
    Order deny,allow
    Deny from all
    Allow from all
</Directory>

Alias /nagios "/var/www/nagios"

<Directory "/var/www/nagios">
    DirectoryIndex index.php
    Options None
    AllowOverride None
    AuthName "Nagios Access"
    AuthType Basic
    AuthUserFile /var/www/etc/nagios/nagios.passwd
    Require valid-user
    Order deny,allow
    Deny from all
    Allow from all
</Directory>

...and my cgi.cfg:

Code: Select all

main_config_file=/etc/nagios/nagios.cfg

(changed this manually:)
#physical_html_path=/nagiosdatadir@
physical_html_path=/nagios
url_html_path=/nagios

(no changes from here down to eof)

At first glance I'd assume a path issue...?

[tmcdonald]

Can you hover over a few of the links in the sidebar and copy the paths? I want to see what is being built.

Also, any errors in your Apache logs?

[mksmr]

This is what the current screen looks like with the provided index.php. I suspect either a path issue or php code not processed properly.

The Apache logs have no suspicious entries.

On hover no popups show up. The first entry in the sidebar ("Curr. Status - Tacticel overview") is a link to
http://<serverip>/nagios/%3C?php%20echo%20$cfg[
(literally what's in the file)


By the way, you will note that with the original out-of-the-box setting, the main page looks like (see below). In my above posting I have attached a screenshot that at least loaded the main.php properly after I had hacked the index.php:

original index.php:

Code: Select all

 <frame src="<?php echo $url; ?>" name="main" frameborder="0" style=""

(this is what you see below)

my hack:

Code: Select all

<frame src="main.php" name="main" frameborder="0" style="">

Probably related, but I have to admit my ignorance about php.

Matthias

[rkennedy]

What is the output of php -V from the CLI? Are you able to verify php is running in conjunction with apache?

Creating a page with <?php phpinfo(); ?> should help provide a bit of information if you're able to do so. Hacking together the functions will work, but it seems like a bigger issue here than just the HTML side bar.

[mksmr]

Aaaalll right (slaps forehead)... we may call this one settled. PHP wasn't enabled in Apache. My fault.

Matthias

[mksmr]

One more thing... I'm still getting a message that nagios wasn't running while it actually is. I understand that the apache user needs to have read access to status files, namely
status.dat
retention.dat
nagios.qh

In this installation, the files are located here:
/var/www/var/nagios/status.dat
/var/www/var/nagios/
/var/nagios/rw/nagios.qh

and they are either group (www) or world readable. I'm clearly still missing something here...

[mcapra]

If it's just a cosmetic thing, I would suggest raising a github issue:
https://github.com/NagiosEnterprises/nagioscore/issues

I can't find anything particularly different about how the pid is assessed between BSD and GNU. Can you share the output of:

Code: Select all

ls -al /var/www/var/nagios

You may also need to update the permissions on the lock file, not 100% certain

[mksmr]

Code: Select all

ls -al /var/www/var/nagios

You may also need to update the permissions on the lock file, not 100% certain

Code: Select all

#  ls -al /var/www/var/nagios
-rw-r--r--  1 _nagios  _nagios  12640 Sep 30 13:17 objects.cache
-rw-------  1 _nagios  _nagios  12989 Oct  4 09:17 retention.dat
drwxrwxr-x  2 _nagios  www        512 Sep 30 13:17 rw
-rw-rw-r--  1 _nagios  _nagios  13423 Oct  4 10:01 status.dat

I changed the ownership/perms on
/var/www/var/nagios/retention.dat to

Code: Select all

-rw-r-----  1 _nagios  www      12989 Oct  4 09:17 retention.dat

just to make sure. No effect, though.

Matthias

[mksmr]

I changed the ownership/perms on
/var/www/var/nagios/retention.dat to

Code: Select all

-rw-r-----  1 _nagios  www      12989 Oct  4 09:17 retention.dat

just to make sure. No effect, though.

Edit - for some reason it's running now. Restart of Nagios brought the old fileperms back on retention.dat, but I can access everything. No idea why it wouldn't do before.

Thanks a lot everybody.

Matthias

[tmcdonald]

Glad to hear it - are we alright to close this thread now?