Nagios Support Forum

Hi;

what is the difference between using :

http://nagiosserver/nagios/jsonquery.html

or

http://nagiosserver/nagiosxi/api/v1/

Is one preferred over the other?

from what I can tell it appears that the first one, jsonquery, seems to have some better capabilities when querying for status(i.e. hostlisting based on groups, hostcount and so on..)

The first one is the JSON CGI for Nagios Core, which while it has some better filtering right now, does not have the ability to do things such as modify the configs, restart core, apply configs, etc. Besides those, most of the same data should show up unless it's XI-specific, which will only show up in the second one.

so for querying, at that moment anyways, the nagios core JSON may be a better option?? sounds like that will change in the future.

I'm wanting to check the health of a number of network switches at a particular site(we have hundreds of sites across the province) and using the nagmap plugin plot the overall group health of the switches at each site for display to management. My plan was to group those switches in hosts groups and then develop a plugin to check the overall group health of each of those groups using a JSON query. Does that make sense or is there an easier way to tackle that??

I just stumbled across the xi api this morning, I do use the \etc\import folder for importing new hosts and services through automation, do the post commands support inserting free variables as well? is there any additional documentation on the XI api outside of what is in the online help?

This presentation has a bit of information - http://www.slideshare.net/nagiosinc/eri ... -nagios/46

On top of that, you can take a look at what's possible through the API by navigating to Backend API URL -> Click 'wrench icon' -> Select the user. From here you'll see all the information that can be used with the API.

More documentation can be found by going to your Nagios XI Home page and clicking 'Help', you'll see a section for 'Nagios XI API Docs' on the left side.

Is that what you were looking for?

paul.jobb wrote:so for querying, at that moment anyways, the nagios core JSON may be a better option?? sounds like that will change in the future.

Yes. There is a bug that will be fixed in the next release of core. Sometimes a complete JSON is not returned. If you come across this, I know it's fixed in core pre 4.1.2 on Github.

paul.jobb wrote:I'm wanting to check the health of a number of network switches at a particular site(we have hundreds of sites across the province) and using the nagmap plugin plot the overall group health of the switches at each site for display to management. My plan was to group those switches in hosts groups and then develop a plugin to check the overall group health of each of those groups using a JSON query. Does that make sense or is there an easier way to tackle that??

BPI might also be an option.
https://support.nagios.com/kb/article.php?id=280

paul.jobb wrote:I just stumbled across the xi api this morning, I do use the \etc\import folder for importing new hosts and services through automation, do the post commands support inserting free variables as well? is there any additional documentation on the XI api outside of what is in the online help?

Basically any directive that can be configured in an object that is defined in the nagios core documentation. However I just did some quick tests and it does not appear these are currently supported. Perhaps define these in a template and "use=" those templates might be a work around.
https://support.nagios.com/kb/article.php?id=356

Some directives are mandatory, however when using templates you might not want to define some directives. You need to use &force=1 to allow that.
NOTE: The "Backend API URL" that @rkennedy referrenced is an older API that works, but is depreciated (it's just a read-only api).

Thank you Troy

force=1 was a good tip thanks for the heads up on that, I don't need free variables for network devices so I can get away without those for now, the following is the powershell I got working this morning with your suggestion.

$body = @{
host_name="AGSGSFL2SW1"
address="255.255.255.255"
alias="AGSGSFL2SW1"
notes="latlng: 51.046259,-114.053315"
_testvar="some string"
use="GOA-Network-Device"
force=1
}

$url="http://nagiosserver.gov.ab.ca/nagiosxi/ ... x&pretty=1"

$result=Invoke-RestMethod -Method Post -Uri $url -Body $body

$result


Thanks also for pointing out BPI, I do use that currently for some other stuff but have sort of moved away from that towards check_multi, BPI is a bit of a chore to maintain and for availability measurement it doesnt seem to filter out soft states which check_multi does.

With the network stuff switches are being added and removed from the environment daily so its just easier to reconcile nagios with there inventory DB through automation scripts a few times a day, group them on location and then measure the availability of the group health of the site .

I really appreciate your input.

Great! Is there anything other questions about the APIs we can help with?

Great to hear you got it working.

paul.jobb wrote:_testvar="some string"

Just curious, did this custom variable make it through into the object, or did the API drop it?

no it didn't make it, hopefully in the next release it does.

FYI, we already have an internal feature request for adding free variables to the REST API (TASK ID 7853).