Monitor web service with Nagios check_http - web-services

Question

I am trying to monitor a web service with check_http plug-in of Nagios.
The url I am trying to monitor includes url parameter.
And it turns out that check_http plugin will ignore the url parameter when checking.
Herewith my configuration.
'check_http' command definition
define command{
command_name check_http
command_line $USER1$/check_http -I $HOSTADDRESS$ $ARG1$
}
define service{
use local-service ; Name of service template to use
host_name pear
service_description HTTP
check_command check_http!-u http://pear.com/total?eId=12345&env=abcde
notifications_enabled 0
}

Answer 1

Try replacing the value passed into -u with a relative path instead of the full URL.
In this example the hostname (-H) will be supplied by $HOSTADDRESS$ which is taken from the address field of the pear host definition.
The value passed into the -u parameter should be a relative path, for example: /total?eId=12345&env=abcde.
We'll add the -u to the check_http_with_args command definition so we don't have to pass it as part of our parameters in our service definition.
define host {
host_name pear
alias pear
address pear.com
use linux-server
contact_groups admins
notification_interval 0
notification_period 24x7
notifications_enabled 1
register 1
}
define command{
command_name check_http_with_args
command_line $USER1$/check_http -H $HOSTADDRESS$ -u $ARG1$
}
define service {
service_description pear_total_http
use generic-service
check_command check_http_with_args!/total?eId=12345&env=abcde
host_name pear
contact_groups admins
notification_interval 0
notification_period 24x7
notifications_enabled 1
flap_detection_enabled 1
register 1
}
In the end, the command that Nagios execute should get translated into something that looks like this:
/usr/local/nagios/libexec/check_http -H pear.com -u /total?eId=12345&env=abcde
You can try executing the above from the command line to make sure it works for you.
Note: Replace the path to check_http with the actual path corresponding to your installation location on your Nagios server.
Relevant parts of the check_http man pages we're referencing:
-H, --hostname=ADDRESS
Host name argument for servers using host headers (virtual host)
Append a port to include it in the header (eg: example.com:5000)
...
-u, --url=PATH
URL to GET or POST (default: /)
Source: https://www.monitoring-plugins.org/doc/man/check_http.html
EDIT:
To answer your questions from the comment, the -k or --header= will allow you to pass in headers.
-k, --header=STRING
Any other tags to be sent in http header. Use multiple times for additional headers
Source: https://www.monitoring-plugins.org/doc/man/check_http.html
So to specify an Accept header, I'd modify the following to look like:
define command{
command_name check_http_with_args
command_line $USER1$/check_http -H $HOSTADDRESS$ -u "$ARG1$" -k "$ARG2$"
}
define service {
service_description pear_total_http
use generic-service
check_command check_http_with_args!/total?eId=12345&env=abcde!Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8
host_name pear
contact_groups admins
notification_interval 0
notification_period 24x7
notifications_enabled 1
flap_detection_enabled 1
register 1
}
...adding -k "$ARG2$" to the command_line of the command definition and adding the Accept: <MIME types> (for example Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8) to the check_command of the service definition.
Also, I wrapped $ARG1$ of -u "$ARG1$" part of the command_line in double quotation marks as I suspect the & in eId=12345&env=abcde is causing the shell to think the end of the command is terminated just before the &. Wrapping the parameter in double quotes should make it see the whole string as one whole argument.