Checkmk Integration (v 1.x)

Use this guide for checkmk v 1.x

With the ilert checkmk notification plugin you can easily integrate checkmk into ilert and extend your existing checkmk system with SMS, push and voice notifications as well as on-call schedules from ilert. The screenshots below are based on the checkmk Raw Edition (CRE) 1.2.8. The ilert checkmk plugin is compatible with all checkmk versions.

Did you know we offer a native integration for checkmk 2.x

System requirements

• Python >= 3.7 (alternatively Python >= 2.7.9, we suggest 2.7.10)

Python 2.x is EOL (end of life) we suggest to use Python >= 3.7

In ilert: Create a Checkmk alert source

1. Go to Alert sources --> Alert sources and click on Create new alert source

   
2. Search for Nagios / Icinga in the search field, click on the Nagios / Icinga tile and click on Next.

   
3. Give your alert source a name, optionally assign teams and click Next.

4. Select an escalation policy by creating a new one or assigning an existing one.

   
5. Select you Alert grouping preference and click Continue setup. You may click Do not group alerts for now and change it later.

   
6. The next page show additional settings such as customer alert templates or notification prioritiy. Click on Finish setup for now.

7. On the final page, an API key and / or webhook URL will be generated that you will need later in this guide.

   

In Checkmk: Configure the ilert plugin

Download the checkmk plugin package and unzip it:

> https://github.com/iLert/ilert-nagios/releases/download/latest/ilert-nagios.zip
> unzip ilert-nagios.zip

For Python >= 2.7.9 support, please use the files in the python2 folder

For checkmk you need the two files ilert_nagios.py and ilert_check_mk.sh. You can delete the remaining files in the directory. Move the plugin file ilert_nagios.py into the directory /usr/local/bin . Both files must be executable by both checkmk and the cron daemon:

> mv ilert_nagios.py /usr/local/bin
> chmod 755 /usr/local/bin/ilert_nagios.py
> chmod 755 ilert_check_mk.sh

Standalone version of checkmk

If you are using the standalone version of checkmk, follow the instructions below. If you are using the OMD version, please go the tab OMD version of checkmk.

> crontab -u nagios -e

Add the following entry:

* * * * * python3 /usr/local/bin/ilert_nagios.py -m send

Via this cron job, events are sent to ilert every minute that failed in the first send attempt (e..g due to a network error).

Move the ilert alerting script to the notifications directory of checkmk:

> mv ilert_check_mk.sh /usr/share/check_mk/notifications

OMD version of checkmk

If you are using the OMD version of checkmk, follow the instructions below.

> su - {site-name}

Create a file named ilert under the cron.d directory of your monitoring instance (e.g. /omd/sites/{site-name}/etc/cron.d). The file should contain the following content:

* * * * * python3 /usr/local/bin/ilert_nagios.py -m send

Execute the following command for the changes to take effect:

> omd reload crontab 
> exit

Move the ilert alerting script to the notifications directory of checkmk:

> mv ilert_check_mk.sh /omd/sites/{site-name}/local/share/check_mk/notifications

Log in to checkmk's web GUI and continue there.

1. Navigate to the "Users" menu and click on "New User".

2. Fill in the fields marked below. Add the user to the contact groups to which the hosts or services that you want to be alerted for. Click on "Save".

3. Go to the notification settings of this user (by clicking on the "Notifications" icon).

4. Click on "New Rule" and select ilert checkmk Plugin as the "Notification Method". Enter your ilert API key (see above) as a parameter.

5. Go back to the user list and activate your changes by clicking on "1 Changes":

Test the integration

You can test the integration directly in checkmk to verify that everything is working. To do this, open a host or service check in the Web GUI and click on the "Execute" icon (Hammer Icon). In "Fake check results" you can manually set the state of a host or service. Click on "Down" (for a host) or "Critical" (for a service) and then confirm with "Yes!". You should now see an alert in ilert.

FAQ

Which notification types are processed by the plugin?

The plugin processes the notification types PROBLEM , ACKNOWLEDGEMENT and RECOVERY. The remaining Notification Types (including FLAPPING* and DOWNTIME*) are ignored.

checkmk has the following notification types:

Types	Description
PROBLEM	Normal host or service problem
RECOVERY	Host / service goes UP / OK again
ACKNOWLEDGMENT	Acknowledgment of a problem
FLAPPINGSTART	A host / service begins to be discontinuous
FLAPPINGSTOP	End of discontinuity
DOWNTIMESTART	Start of scheduled maintenance.
DOWNTIMEEND	Normal end of maintenance
DOWNTIMECANCELLED	Premature termination of maintenance
CUSTOM	Alarm triggered manually by command
ALERT HANDLER	Alerthandler execution (CEE from 1.4.0i2)

What happens if my internet connection is lost? Are the events generated in checkmk lost?

No, events won't be lost. The plugin stores the events locally in a temporary directory (by default in /tmp/ilert_nagios) and tries to send them to ilert every minute. This means that as soon as your connection is available again, cached events will be sent to ilert. In addition, we recommend that you monitor your Internet connection using our uptime monitoring feature.

Will alerts in ilert be resolved automatically?

Yes, as soon as the state of a host / service is UP or OK again in checkmk, the associated alert is resolved in ilert. If a problem is acknowledged in checkmk, the associated alert in ilert is set to the status Accepted.

Can I link checkmk to multiple alert sources in ilert?

Yes, create a checkmk user for each alert source in checkmk. Proceed as described above in the instructions.

The plugin does not work. How do I find the issue?

Please look first in the log file. The plugin uses the Unix / Linux system log for logging (eg under /var/log/messages or /var/log/syslog ). If you can not find the error, please contact our support at [e-mail] .

Please also have a look at the checkmk documentation, where additional log files related to alerting are documented: The course of a notification from beginning to end