disk monitoring queue length etc...

[Peter Welter <peter.welter (at) gmail.com>]

I've searched many times on this list 'How to set up monitoring disk drives
in Windows'. However, I've never found a real solution. So, we have done
some scripting ourselved querying WMI and want to trend this output.

If I use the Totals (for all the disks) it is pretty straight forward. This
works as expected.

However, if I would like to trend to a specific disk, it becomes more
complex.

My test for disk C produces the following:

_DskQL:0
_DskRdQL:0
_DskWrQL:0
_DskBtsPerRd:17423
_DskBtsPerTr:11933
_DskBtsPerWr:11246

Now, what I would like to have is a rrd.file for each drive letter, in this
case drive_c.rrd, (just like the disk-test, all the disks are shown in one
graph but each disk has an rrd-file) but I do not know how to continue.
Drives under Windows can have any letter (C, D, ...Z) and I do not want to
fill in the entire alphabet in my hobbitserver.cfg, but perhaps this is the
only way to go without real programming.

But perhaps data trending, instead of status, would be a better approach.

But how should I do this? Any help much appreciated. If this works, I will
post the result on Xymonton.

Regards,
Peter