Aug 30, 2012 10:24 AM
Logical volumes issue v 4.2.0 zenoss
-
Like (0)
Hi,
there is a little issue with logical volumes when upgrading to version 4.2.0.
Component 'File Systems', if there are several logical volumes, zenoss don't succeed to count
Used Bytes, Free Bytes, %Util
Device: Device
Component: zen.winperf.PerfRpc
Severity: Warning
Time: 2012/08/30 17:04:24.000
Message:
Bad counter for device 192.168.11.46: \LogicalDisk(\\?\Volume{69126123-19da-11e0-92d2-806e6f6e6963})\Free Megabytes
....
Bad counter for device 192.168.11.46: \LogicalDisk(\\?\Volume{69126123-19da-11e0-92d2-806e6f6e6963})\Disk Write Bytes/sec
.....
Bad counter for device 192.168.11.46: \LogicalDisk(\\?\Volume{69126123-19da-11e0-92d2-806e6f6e6963})\Disk Read Bytes/sec
Does anybody has the same issue, or are there any resolve for it.
I'm not getting the error messages you are, but I'm ceratainly not getting any data from WMI about any of my Windows servers volumes.
Hi,
I have the same issue with the system-partitio (this 100 MB thing)...
But no hint except disable monitoring. But modeling the device every 12 hours, enables the check again :-(
Bernhard
Have the same problems ... :-(
Look at this about how to get rid of the alerts
Mentions a bit 2/3 down about a transform on the event which seems to work OK for me on 4.2
Question I've got is that I've got a clustered shared volume on a Hyper-v cluster which returns the same sort of missing data, problem here is that it's a real volume not the crappy 100mb hidden partition and I really really need to see the % usage on it. It never gets a drive letter though as it's effictively mounted as a junction point.,
Any clever ideas?
We have built a quick python code fix for this in the FileSystemMap module...
this might help (might.. )
Just tried this and it does get rid of the annoying 100mb partition but doesn't help my main problem.
Cluster shared volumes only appear with the //?/ name so this gets rid of those as well.
Good work though, I realise I'm in the minority using Hyper-v not VMWare
We have this same problem with some special, non-lettered drives. We decided to get rid of these special drive zenoss WMI components altogether in the Windows WMI modelling because the OS currently does not provide any framework for WMI, perfmon or SNMP to gather any reasonable info from them. Most of the current storage units we are using have pretty good SNMP or SSH onboard which provide much better "from the source" information about the current state of the arrays/volumes - so we decided to scrap any modeling through windows and get it from the storage units themselves.
In your current setup i dont think there is any way you will be able to get WMI modeled stats from your drives as the volume connections are through proprietary drivers which have no perfmon counters included.
Follow Us On Twitter »
|
Latest from the Zenoss Blog » | Community | Products | Services Resources | Customers Partners | About Us | ||
Copyright © 2005-2011 Zenoss, Inc.
|
||||||||