# 

How to use the JSON interface to Check_MK is described at 
http://mathias-kettner.de/checkmk_multisite_automation.html

it also shows the URL format with the following example:
curl 'http://localhost/heute/check_mk/view.py?view_name=svcproblems
&output_format=JSON&_username=auto&_secret=BLARBS@JQTAH@UWMBJXB'


Modifying this to give the data for only one host - normally you need to 
specify a site name but we don't actually know it. we can skip it as long
you don't have duplicate host names between sites. And I really hope you don't

curl 'http://OMD-server/omdsite/check_mk/view.py?host=myhost&view_name=host&output_format=JSON&_username=auto&_secret=myauthkey'

Starting from that query (which the python script also uses) you can select
data you want to be displayed.


An example would be to use put all critical software metrics (full disk, 
network interface errors, missing backup...) into a single servicegroup and then
change the view you're querying to one that shows this servicegroup.

You would query the service group, your current host and filter for errors.
That way you'll be able to tell if there is a critical miscondition on your
system.

You can parse that output right now and later process it in your own scripts.
Another idea is to have exit code handling for cmkcli and then you would 
just use it like 

script_that_prepares_something
cmkcli && 
script_that_runs_something

That way your second script would only run if the Nagios status of the host
is "good to go".


# TODO
1. add cli return codes
2. Just review add comments to all functions 
3. add another output
4. add setup
5. cover with tests




# HOW TO GET OMD instead of doing old ages' Nagios installs:


Here you can download OMD - which is the best way to install Nagios
Pick the right version for your distro.


http://labs.consol.de/lang/de/nagios/omd-repository/

You'll want a recent version since there's an issue with the JSON data in
older versions.

for centos/rhel you need the epel or similar repos installed

for debian ubuntu just install....

You _ALSO_ must switch to apache mpm prefork
(aptitude search ...)


after installing do


	# create two fullblown nagios envs
	omd create mytestsite
	omd create mytestsite2

	# launch your two new nagios instances
	omd start


	and follow the instructions there