[Nagios-devel] Feature request

Anthony Mendoza anthony at onetruemedia.com
Fri Jan 12 19:18:27 UTC 2007

Something I've used in the past specific to SMART checking is to use
SNMP and wrap the smartctl -H /dev/had (forgot exact syntax, but run the
health check) and then return the status via SNMP.  This would allow you
to do active checks of your smart deamons.  I didn't read the whole
thread so I apologize if you already mentioned that active checks or
SNMP isn't something you wanted to do.

I've had to wrap many commands via SNMP that weren't available via
standard OIDs or were only available via NRPE since I don't use NRPE.

Hope this help some.

> -----Original Message-----
> From: nagios-devel-bounces at lists.sourceforge.net 
> [mailto:nagios-devel-bounces at lists.sourceforge.net] On Behalf 
> Of Horvath Tamas
> Sent: Friday, January 12, 2007 11:09 AM
> To: nagios-devel at lists.sourceforge.net
> Subject: Re: [Nagios-devel] Feature request
> Hello!
> Thanks you all of them to write to me ideas.
> I have read the documentation and I use freshness checking, volatile
> services and other features according to the documentation 
> when they fits my
> need.
> But as Thomas Guyot-Sionnest wrote: smartd only send alerts 
> when a disk goes
> bad (or when the daemon is started).
> "The purpose of "freshness" checking is to ensure that 
> service checks are
> being provided passively by external applications on a regular basis."
> So "freshness" checking does not fit for my scenario.
> "Volatile services are useful for monitoring...
>     * things that automatically reset themselves to an "OK" 
> state each time
> they are checked
>     * events such as security alerts which require attention 
> every time
> there is a problem (and not just the first time)"
> So this does not fit for my scenario.
> I know that I have to be sure that my passive service setup 
> is not stale as
> Chris wrote.
> So I have two options:
> 1. Periodically restart smartd to send a test message to me 
> every restart.
> But in this case my non ok status would change to OK. I do 
> not want it to
> happens.
> 2. Use another check to see, whether smartd process is 
> running on host. I do
> this. But this does not solve my first problem.:(
> It solved the idea by Nicolas COLIN. Sorry I always forget external
> commands, becaue I use them very rarely:) Now I know what a 
> silly question
> was this. I'm sorry...
> I think Thomas Guyot-Sionnest's configuration is very 
> interesting. I will
> use it for an other passive check type.
> Thanks again all of you.
> Regards, Tamas

More information about the Nagios-devel mailing list