URL monitoring no graph

[RebeccaIlene]

Hi,

We have set up a monitoring for URL externally so that the connectivity for the URL is checked on an external host.
This is defined using commands and then the output form that command is called by Nagios XI and displayed as the final output.
This helps us to understand the connectivity from an external server to internal URL.
The issue is that we are not able to view the performance graph for these services.

Can someone please help with this?

Regards,
Rebecca

[lmiltchev]

Not all of the checks return perfdata. Does your check return perfdata? To find out, click on your service under the "Service Status" page, then click on the "Advanced" tab. Do you see anything next to the "Performance Data" field? Can you show us a screenshot of this page?

[RebeccaIlene]

Hi,

Please find attached screenshots of example 1. Will post again with another example.

[RebeccaIlene]

Here is another example.

[lmiltchev]

These are screenshots from the Overview tab, NOT from the Advanced tab...

First, you need to see if your check actually returns performance data by going to the Advanced tab. See example below:

example01.PNG

If your check doesn't return perfdata, there is nothing you can do about it. Well, if you had a developer, you could modify the plugin, in order to make it return perfdata. You could follow these guidelines:

https://nagios-plugins.org/doc/guidelines.html#AEN200

Alternatively, you could try using a different plugin that returns perfdata.

If your plugin returns perfdata (you see some values under the Advanced tab), but you are still not seeing graphs, you would need to troubleshoot the issue, by following the steps, outlined in the KB article below:

https://support.nagios.com/kb/article/n ... ems-9.html

[RebeccaIlene]

Hi,

Thanks for your help with this.

Attached are screenshots of the advance tabs of the services.

Have performed the steps mentioned in the link provided, but still no difference.

Please help with this.

Thank you,
Rebecca

[lmiltchev]

The perfdata from the "Procs" check should look like the screenshot I posted. Yours has an extra pipe ("|") in the output. This is NOT a correct perfdata output. According to the Nagios Plugin Development Guidelines, the perfdata should look like this:

Code: Select all

'label'=value[UOM];[warn];[crit];[min];[max]

Note that there is no pipe in the output.

https://nagios-plugins.org/doc/guidelines.html#AEN200

What is the plugin that produces this output? Can you show us the actual check run from the command line, along with the output of it?

The second screenshot you showed us looks fine. Aren't you seeing graphs for this check? You can try removing the "old" perfdata by going to "/usr/local/nagios/share/perfdata/<hostname>" and deleting the RRD and XML files for this particular service:

Code: Select all

rm -rf <service_description.xml> <service_description.rrd>

The "old" perfdata (for this specific service only) will be lost. However, the deleted files will be recreated, and you can check in 15-20 minutes to see if the graph start showing up in the GUI.

[RebeccaIlene]

Thanks for your reply.

Have tried deleting the RRD and XML files as suggested. The files got recreated but the graphs still do not show any data. Not sure, if this is because the plugin that we are using to get the output is not check_http. Please suggest on this, if you want me to share the plugin that we use.

Also, please find attached the command that we are using to check the total processes.

Let me know if any other input is required to investigate this.

[lmiltchev]

Please suggest on this, if you want me to share the plugin that we use.

Sure. Please upload the plugin that you are using on the forum. Change the extension of the file to "txt" if needed (not sure what type of files are allowed for upload).

The check_procs.sh plugin that you are using outputs two pipes, which is most likely causing the issue. Why don't you use our official check_procs plugin instead? Run the command below and show the output:

Code: Select all

/usr/local/nagios/libexec/check_procs  -w 400 -c 500

You should see only one pipe in the output.

Example:

Code: Select all

PROCS OK: 181 processes | procs=181;400;500;0;

Using the "check_procs" plugin instead of "check_procs.sh" in your service definition should fix the issue. You still will have to delete the "old" RRD and XML files for this service and wait for them to be recreated (with the new plugin).

[RebeccaIlene]

Thanks. Please find attached plugin that we are using. We have defined the check_http commands on the server where the services are hosted and using this plugin we call the command to give the output on Nagios. Let me know if this does not make sense.