Can I monitor a host without defining any services for it?

[Ian Masters]

Thomas

Thanks for the reply.

>>> You can use check_dummy for the service... In Nagios 3 with regularly
>>> scheduled hosts checks it should work.
>> Well, at the moment I only have about 20 hosts being tested, and even
>> though I have check_interval set to 1 in the host definition, looking at
>> the scheduling queue, I can see that host checks are only occurring
>> every 80 seconds. When I've added about 200 more hosts I'm wondering if
>> this time lag will increase.
>>
>> If that is the case, then I guess I will have to use check_ping after all.
> 
> I'm not sure what would cause that... AFAIK in Nagios 3 host checks end
> up in the same queue as service checks

Looking at the scheduling queue, it certainly looks like it ...

>, so the only explaination I could
> find would be:
> 
> 1. Your interval_length is larger than the default of 60

It is set to 60. I tried setting it to 30, but that also produced
unreliable results in the scheduling queue.

> 2. check_result_reaper_frequency is too large (will delay the results,
> and therefore delay the time at which nagios will re-schedule the checks)

Again, I haven't changed the defult value of 10.

> 3. Various commands (notifications, event handlers, performance data)
> slowing down nagios (those are serialized just like host checks in
> Nagios 2 - see below)

Not sure about that one ... at the moment my setup is pretty small (18
hosts and 26 service checks) and all notifications are turned off (we
only require sound alerts)

> The logic in Nagios 2 is to check host only when a service fail, and it
> does not support assync host checks (only service checks are assync).
> The problem with it is that the host checks are serialized, so it adds
> an artificial limit at which they can be performed (the time it takes to
> detect a host down). This was an annoying bottleneck on huge setups.

OK, if host checks are serial, then I guess it would make sense for an
unknown amount of time to be added.

>> According to check_dummy -h:
>> This plugin will simply return the state corresponding to the numeric
>> value of the <state> argument with optional text
>>
>> Am I right in thinking that this does not actually perform a test? It
>> just returns a state (OK, Warning etc) depending on the argument you
>> give to check_dummy. Is that correct?
> 
> Right. I mainly use it to make passive check with freshness checking
> send an alert when the check becomes stale, and to reset volatile
> services automatically.

That explanation is a little advanced for me right now. Still at a very
early testing stage.

>> If that is the case, is there really any point in using it for my
>> purposes. Obviously it prevents the Nagios warnings, but other than that
>> is it really serving any purpose?
> 
> If it's only to avoid a warning, then you should just live with it. I
> never tested myself so I assumed it wasn't working at all.

But if the only way I can be sure that Nagios will check the hosts that
only require a ping every minute is to add ping as a service, I guess
that's what I'll have to do. Slightly frustrating but I'm sure we'll
learn to live with it.

Thanks a lot.

Ian

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Nagios-users mailing list
Nagios-users at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. 
::: Messages without supporting info will risk being sent to /dev/null