Nagios Support Forum

Our network team is having an ongoing issue with a provider and has asked us to set up nagios monitoring of a couple of external sites to check on network performance.
They want us to ping every 5 seconds.

I found this viewtopic.php?t=50557

And after reading set this up:

Looks to be working as advertised – or at least trying.

For thresholds to set they want the following:

>> I would like it if we can get a pings every 5 seconds and if packet loss is over 10% over 5 minutes to alert?
Also if latency is higher than 100ms for 5 mins to email alert.

I want to know the following:
• what should I use for thresholds for check_icmp ?
• They don’t want a warning – just the critical. Is it possible to omit the warning all together
• This is set up as an active check we don’t use passive checks. What I have read on enabling freshness concerns passive checks.
Does it work properly with active checks?

Thanks for you help,
Penny Karr
[email protected]
Atrius Health

Hello @pkarr,

If you want an alert for latency above 100ms or packet loss above 10%, you can set your thresholds for check_icmp to -c 100.0,10%.

Your Nagios Core and Nagios XI configuration will need some modifications, though.
Nagios Core is built to check every minute and will require some configuration modification if you would like for it to check every 5 seconds. You will need to reduce the Nagios Core interval_length from 60 to 5 to allow Nagios Core to process checks every 5 seconds instead of the default of every minute. This is not the intended use of Nagios Core and it may not work properly if you do this, particularly if you have a lot of checks.

As for your XI configuration, freshness is not relevant for active checks. You will need to update your check interval, retry interval and max check attempts to check every 5 seconds, retry every 5 seconds and to have the number of attempts checked to be equal to 5 minutes (12 5-second intervals per minute, so 12 intervals * 5 minutes = 60 check attempts).

As noted above, this is not the intended use for Nagios Core/XI and you may be better off using passive checks.

pkarr wrote: ↑Wed Jun 05, 2024 1:55 pm Our network team is having an ongoing issue with a provider and has asked us to set up nagios monitoring of a couple of external sites to check on network performance.
They want us to ping every 5 seconds.

I found this viewtopic.php?t=50557

And after reading set this up:
host_template_ext_site_freshness_on.jpg


Looks to be working as advertised – or at least trying.

For thresholds to set they want the following:

>> I would like it if we can get a pings every 5 seconds and if packet loss is over 10% over 5 minutes to alert?
Also if latency is higher than 100ms for 5 mins to email alert.

I want to know the following:
• what should I use for thresholds for check_icmp ?
• They don’t want a warning – just the critical. Is it possible to omit the warning all together
• This is set up as an active check we don’t use passive checks. What I have read on enabling freshness concerns passive checks.
Does it work properly with active checks?

Thanks for you help,
Penny Karr
[email protected]
Atrius Health

Hey @pkarr

Thanks for reaching out and hopefully we can help clear things up.
1. When you say thresholds, which thresholds are you talking about? If you are talking about which arguements to put in for critical and warning thresholds then from the documentation youll want -c to be 100,10%. Flap thresholds dont sound like they matter in your situation if they only want to be notified on a critical, and freshness threshold is more for active checks.
2. If you only want to be notified for criticals you can modify the script or make the warning value much higher than the critical that should never go off. i.e. 100000,100%.
3. Like you mentioned freshness checks are mainly for passive checks. Since active checks come in at a certain interval you know when the last check was however with passive the freshness checks are to make sure that a check may still be valid.

These docs may be relevant if you want to read up more on this.
https://nagios-plugins.org/doc/man/check_icmp.html
https://assets.nagios.com/downloads/nag ... hness.html