Windows Monitoring

[transmax]

I've done all the google-fu I can, and I seem to have failed.

I want to monitor average seconds per disk transfer. Perfmon has a counter for this, and that's good.

Unfortunately, check_nt only gets a value for the instant it queries the windows server, not an average over some length of time (say 60 secs).

Any clues on how to overcome this so that I can catch when the server has a sustained problem over a minute or so?

BTW, this actually applies to many perfmon counters as far as I can tell.

Thanks for any help you can give.

Cheers,

Lee Benson
Transmax

[mguthrie]

Typically this is done with the following plugin, but it doesn't look like is has parameters other than true/false for checking counter averages:
http://www.nsclient.org/nscp/wiki/CheckCounter

You could also check to see if this data would be available via SNMP, but SNMP can be a ugly thing to work with...

[transmax]

CheckCounter seems to work the same way, and actually I think check_nt just invokes checkcounter at the target.

From your reply, however, it seems unlikely than a plugin exists. I was hoping to avoid running scripts on the target but that might be the only way.

Regards,

Lee

[scottwilkerson]

I think you are right, to get a longer sample you are going to need to poll the data at an interval and have a script do the math.