Host state and response details

[--JayJay--]

I have configured a Nagios (version 3) monitoring solution for a few hosts, with exim4 email notification on host down, unreachable, recovery state.
The nagios server is syslog server as well, for some important hosts to be monitored.
The nagios configuration is tweaked in the sense that nagios is sending the check_host_alive command to some hosts every 10 seconds.

So far so good.

I am missing a detailed view of the hosts being pinged every 10 seconds, and missing their response as well.

Is it possible to have this granularity, this "zoom functionality", with nagios 3 , out of the box?
Or can this be done with some add on , plugin? If so....which ones?

As I can 't get more detail out of nagios , than you can see in attached picture.
I would like to be able to see every command nagios sends, and every response -if any- as well........

[slansing]

Well, I believe the most granular way to view data returned from plugins in the web interface would be through Home > Hosts/Services (then clicking on your object to get to the details page) The information shown should be exactly what is being returned by the plugin, so you shouldn't be missing out on any data that has actually been returned to Nagios.

[--JayJay--]

When Nagios (or the plugin) is pinging a host every 10 seconds, and gets an ICMP reply (or not).
Aren't these data/events logged one by one in a database?

[eloyd]

Generally speaking, Nagios only logs state changes. So if you want the specifics of a particular check, you can turn up some debugging in the nagios.cfg and you will see the individual checks (and results) in the nagios.log.

[tmcdonald]

JayJay, were you able to get the logging tweaked as per eloyd's suggestion?

[--JayJay--]

debugging indeed generates the lines I wanted to see.
But it also generates load, which is substantial as I am running the nagios Core on a raspberrypi with limited resources.

I use fping in combination with tcpdump on a different Pi to record and store the ping and echo.

[eloyd]

Debugging is usually only used for debugging. Is there a reason you need to see every result rather than just state changes? And can you go longer than 10 seconds between pings?

[--JayJay--]

Is there a reason you need to see every result rather than just state changes? And can you go longer than 10 seconds between pings?

The nagios system is running in soho environment with frequent internet connection flapping, i.e. the internet router is rebooting at certain moments.

-the fping/tcpdump logfile is another source, besides nagios state changes, in case the internet connection is flapping.
So , maybe not strictly necessary, but a "second opinion" in case things go wrong.

-ping frequency could be lower than every 10 scnds., could be decreased to something between 30 and 60 seconds.
The ping frequency must be higher than it takes the internet router to reboot. Rule of thumb is that it can do that in less than 1 minute.

[eloyd]

I'm curious why you're monitoring this so aggressively, though. And are you monitoring it from inside the environment or from the outside? The reason I ask is that, by understanding what you're trying to accomplish overall, we may be able to help you do things a more efficient way.

[abrist]

I agree. A bit more information concerning the use case would help.