Hosts w/o services

[Jay R. Ashworth]

On Wed, Apr 16, 2008 at 11:20:29AM -0800, Israel Brewster wrote:
> To some extent, yes, but not necessarily a service we can monitor.  

There are cases that fit this description, but I'm not sure if the
examples you provide are.

> There are quite a few devices on the network that it is important to  
> us to know should the device fail, but aren't offering any separately  
> monitorable "service", per se. For example, printers. We need to know  
> if one dies, so we can fix it before it becomes a problem (not to  
> mention that fixing the boss's printer before he even knows it has a  
> problem makes you look good ;-) ), but other than the simple ping host  
> check, they don't offer any "services" we can monitor.

Telnet to 9100 and look for a banner.  Or the equivalent for non
JetDirect printers..

> Even some more essential devices such as switches can fall into this  
> category, as they are just routing traffic, not offering any  
> "services". I could, of course, go crazy and use check_snmp or  
> something to monitor each port on said switch as a service, but that  
> is way overkill for our needs- we just need to know that the switch is  
> there and functioning, i.e. host check.

And moving traffic; you might find it useful to ping-check other things
on that switch.  Or at least ssh-check the switch controller itself.

> Not to mention the rather large category of client machines, which we  
> need to know are running so they can be backed up, and, of course,  
> used for whatever the user needs. Kind of difficult to monitor the  
> ability to run office, or an e-mail client :-). We still want to  
> monitor the host, though, so that hopefully if/when a client machine  
> should die we can fix it before the user (who may well come in before  
> us in the morning) is impacted. We could, of course, monitor something  
> like ssh on those machines, but why? All we (and the user) care about  
> is that the machine is functioning.

Sure.  But for workstations, you're not monitoring diskspace?  Open TCP
listens (to watch for trojans)

> So yeah, while this may not be the way nagios is designed to work, and  
> may never be (which I can live with if so), I really don't see this as  
> being all that unusual a situation, as some responses seem to imply.

Well, it's not that it's unusual, I think; I believe the assertion
being made is that only doing a ping is not the Best Practice.

Cheers,
-- jra
-- 
Jay R. Ashworth                                                jra at baylink.com
Designer                +-Internetworking------+---------+            RFC 2100
Ashworth & Associates   |  Best Practices Wiki |         |             '87 e24
St Petersburg FL USA    +-http://bestpractices.wikia.com-+     +1 727 647 1274

      If you can read this... thank a system administrator.  Or two.  --me

-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference 
Don't miss this year's exciting event. There's still time to save $100. 
Use priority code J8TL2D2. 
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
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