Nagios Support Forum

I didn't want to hijack this thread:
http://support.nagios.com/forum/viewtop ... 086#p98089

but I have the same problem. I was able to get the checks to pass in CCM, but the automatic or forced checks in XI still fail. I did a "Save" and "Apply Configuration" in CCM before scheduling an immediate check.

COMMAND: /usr/local/nagios/libexec/check_nt -H server.fqdn -s "password" -p 12489 -v SERVICESTATE -l "MSSQL\$BLACKBERRY" -d SHOWALL
OUTPUT: MSSQL$BLACKBERRY: Started

But in the Nagios XI interface it still shows: Critical MSSQL: Not found
The "MSSQL" part listed above changes to be the first part of the service shortname before the $ in Windows Services applet.

ie for the Lync server:
COMMAND: /usr/local/nagios/libexec/check_nt -H server.fqdn -s "password" -p 12489 -v SERVICESTATE -l "MSSQLFDLauncher\$PANAMLYNCARC" -d SHOWALL
OUTPUT: MSSQLFDLauncher$PANAMLYNCARC: Started

but the XI interface shows:
Critical MSSQLFDLauncher: Not found

Anyone have an idea? I have 24 of these checks that all went to "WARNING" after the upgrade and they just show "MSSQL\: Error" in the XI interface. None worked as-is in CCM till I made the changes suggested in the other forum thread. Making the changes to the \$ got the CCM checks to pass, but now all are Critical in XI.

old version: 2012R2.9 <-- Everything worked
Upgraded to: 2014R1.1 <-- All MSSQL Service checks with $ in the service shortname show WARNING/CRITICAL depending on changes made to CCM.

From Admin|System Profile:
System:
Nagios XI Version : 2014R1.1
server.fqdn 2.6.18-348.16.1.el5 x86_64
CentOS release 5.9 (Final)
Gnome Installed
Apache Information
PHP Version: 5.1.6
Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:29.0) Gecko/20100101 Firefox/29.0
Server Name: <nagios server>
Server Address: <IP>
Server Port: 80

There is a disconnect between the CCM test check and the way the check is actually run by XI. The test check option adds escaping, which works fine for many checks, but not all checks. For your specific case, change the escape so that the actual service check works, while understanding that this specific check will not work with the "test check command" feature in the ccm.

Confirmed. Setting up CCM to use:
MSSQL\$BLACKBERRY

causes CCM to report:
OUTPUT: Name or service not known
could not fetch information from server

But XI does show:
OK: All services are in their appropriate state.

Weird.
Thanks for your help. You can close/lock this if you wish.

Yep, as abrist mentioned, it is because of how apache takes the test check and executes it literally, going to go ahead and lock this up, let us know if you need it re-opened.