[Nagiosplug-devel] NSclient and Physicaldisk-bytes/sec

Michael Markstaller mm at elabnet.de
Tue May 6 06:09:07 CEST 2003


Hi,

I'm using NSclient (1.0.7.1) and check_nt (nagios-plugins 1.3.0 / 1.7) in conjunction with a shell script to pull some W2k-perfcounters from several servers into MRTG which work fine so far for Memory,SMTP-Messages,PagePool etc. Just to point out it's working overall fine.. Also using from within Nagios to monitor works fine.
Now I've added the following counters which return me somewhat strange values:
"\\PhysicalDisk(_Total)\\Avg. Disk Queue Length" : polled with check_nt it's *always* 0, in perfmon it isn't.

"\\PhysicalDisk(_Total)\\Disk Read Bytes/sec" / "\\PhysicalDisk(_Total)\\Disk Write Bytes/sec": 
the values I receive are very "static" and *very* strange. In perfmon I see a normal graph showing some movement, so the values within perfmon are correct. An example to illustrate:

Values from Perfmon "\\PhysicalDisk(_Total)\\Disk Write Bytes/sec"

197562,763446308	15337,7748891304	37833,0113271332	227493,179856569	225081,310079066	197605,281148608	14067,0434390914	12074,8110884988	185230,215176143	312124,651960133	128784,648122873	104239,2056829	93099,405679069	158402,832409301	320800,643613903	225597,324605396	81592,1992627104

So they're variying as I'd expect.


when using check_nt to poll via NSClient from the same server I get (only changing a few bytes when monitoring for several hours):
169915,169914,169913,169912,169912,169911,169910,169909,169909,169908,169907,169906,169905,169905,169904,169903

Also, when there comes load to the disks, nothing is changing on these counters..

I've tested with different servers and some other counters on the same machines work fine, anybody has an idea what this is caused by ?


thanks,

Michael




More information about the Devel mailing list