locked
Logical Disk Free Space Monitor Stopped Working with Upgrade to Windows Server Management Pack v6.0.6958 RRS feed

  • Question

  • I upgraded a SCOM 2007 SP1 installation last week to utilize the newly released Windows Server Management Pack v6.0.6958 (released 10/18/11) and the Logical Disk Free Space Monitor stopped working/firing.

    Looking at the Monitor, everything looks fine like it has been for almost 18 months prior to upgrading the Windows Server Management Pack.  Has anyone had this happen and what is the solution for fixing it?


    Bill Thacker
    Thursday, November 17, 2011 3:57 PM

Answers

  • I never got help for resolving this via the TechNet Forum, so I had to open a case with Microsoft Professional Support. We ended up having to implement a workaround (custom monitor) that still can't extract the free space amount in MBs although it does provide the percent free.

    Microsoft acknowledged that this was a bug/deficiency in the product (management pack), so they didn't charge me for the incident. I'm now waiting patiently for a revision to the Windows Server management pack that will address this issue.


    Bill Thacker
    • Marked as answer by Bill Thacker Friday, December 30, 2011 4:44 PM
    Friday, December 30, 2011 4:44 PM

All replies

  • Hi,

    Is there any errot in your event log ?


    Regards, Samir Farhat Infrastructure Consultant
    Thursday, November 17, 2011 4:32 PM
  • Nothing odd within the Event Logs that I can tell.  The Monitor just doesn't appear to work/fire anymore and I really need to get this working again because I have numerous servers with low disk space issues (I'm flying blind).
    Bill Thacker
    Thursday, November 17, 2011 5:12 PM
  • any chance you've disabled the disk counter (cmd > diskperf)?

    As the monitor uses a script (located somewhere in the install dir of an agent), you can try and run this manually to troubleshoot.


    Rob Korving
    http://jama00.wordpress.com/
    Thursday, November 17, 2011 8:02 PM
  • Not intentionally or to my knowledge (seriously doubt it).

    Can you provide more detail with regard to how I would troubleshoot this in order to rule it out?


    Bill Thacker
    Thursday, November 17, 2011 9:55 PM
  • Hi,

    You can review the know issus for this MP in Kevin Holman's blog: http://blogs.technet.com/b/kevinholman/archive/2011/09/30/opsmgr-new-base-os-mp-6-0-6956-0-adds-cluster-shared-volume-monitoring-bpa-and-many-changes.aspx

    The “Logical Disk Free Space” monitors were completely re-written.  The datasource and monitortype was changed from a script that runs once per hour and drives monitor state, to a new script that runs once every 15 minutes, and drives monitor state after 4 consecutive samples.  That seems like a good design change to control any noise from fluctuating disks.  However, running the script every 15 minutes might increase the performance impact with more scripts per hour executing on your agents.  The script datasource no longer outputs the %Free and MBFree values in the propertybag, therefore – these had to be removed from the Alert Description and Health Explorer.  The monitor still works as designed – it creates an alert whenever the threshold is breached.  The only change exposed to the end user – is that these values for actual free space in MB and % are not going to be exposed to the alert notification recipient.

    Regards.


    Carlos Suárez
    Friday, November 18, 2011 12:10 PM
  • Carlos- Thanks so much for providing the link to Kevin's blog (very informative).

    You mentioned that the monitor still creates alerts based on the defined criteria (I had/have overrides for this criteria- could this be causing the problem?), but that is no longer the case for me.  The Monitor just doesn't appear to work/fire anymore and I really need to get this working again ASAP because I have numerous servers with low disk space issues (I'm flying blind which is not cool).

    I'm probably a 3-4 out of 5 when it comes to SCOM 2007 SP1 "internals", so I need a little help with the steps involved for troubleshooting this broken Monitor.  Can ypu assist here and/or point me to someone who can?

    Thanks again!


    Bill Thacker
    Friday, November 18, 2011 3:21 PM
  • Bill, I have not heard of any reports that the new MP is not working.

    One thing is CRITICAL - make SURE you are using 6.0.6958.0 and not the interim release which was 6.0.6957.0.  If you ever imported 6.0.6957.0, and already had existing overrides, this would cause these free space monitors to unload, due to an override param on the monitor itself that was incorrect (misspelled). 

    It is possible if you ever imported this 6957 version, then the monitors unloaded, then upgrade to the new version 6958, the monitors might stay unloaded until the healthservices are restarted on the agents.  I am not 100% sure.  The free space monitors in 6958 work fine in all instances I have seen.  Can you take one of you affected systems, bounce the healthservice, then watch the OpsMgr event log for information about ANY workflows that unload (will be a warning or critical event) and report back?


    Kevin Holman http://blogs.technet.com/b/kevinholman
    Friday, November 18, 2011 3:57 PM
  • Not saying you need to do this - but another option is to implement my addendum MP, which disables these new monitors and implements the old ones again.  However, you would have to re-setup any overrides you have in place, and I'd recommend sealing these MP's I shipped with your own key.

    http://blogs.technet.com/b/kevinholman/archive/2011/11/17/opsmgr-logical-disk-free-space-alerts-don-t-show-percent-and-mb-free-values-in-the-alert-description.aspx

     

     

     


    Kevin Holman http://blogs.technet.com/b/kevinholman
    Friday, November 18, 2011 3:58 PM
  • Kevin- I only import/upgrade MPs that are published through the SC Marketplace, so this was not the case in this scenario. The version of the OS MP that was installed previously was the one from early 2010, so I never knew about or messed with any interim released MPs.

    I've patched and restarted numerous computers in the past 2 weeks since upgrading the OS MP, so I guess I could check their event logs. What exactly am I looking for with regard to workflows that unload?


    Bill Thacker
    Friday, November 18, 2011 4:46 PM
  • Let's save this one as a possible Plan B if I can't resolve the core issue. Thanks.
    Bill Thacker
    Friday, November 18, 2011 4:47 PM
  • Kevin- I just checked one of my servers (Win2003 R2 x86) in scope that has a low free disk space condition (1.67 GB free out of 19.9 GB for C: volume which used to trigger an alert for my overridden thresholds defined).

    I patched and restarted it Wednesday evening and the Operations Manager Event Log is squeaky clean since then (no Errors or Warnings). I have a feeling this issue will be easy to "fix", but I don't know exactly where or what to look for.

    Thanks again for your help.


    Bill Thacker
    Friday, November 18, 2011 7:22 PM
  • Kevin- I still need help with the steps involved for troubleshooting this broken Monitor (needs to be resolved ASAP as I have no visibility into the free disk space situation affecting 165 servers). Can you assist here and/or point me to someone who can? I'd like to avoid spending $250+ with Microsoft Professional Support (see below), but I guess I will have no choice if no one can navigate me through this.

    1) I shouldn't have to pay for a MP upgrade issue

    2) It's usually 50/50 if you get assigned someone half way around the world that you can actually understand (language barrier).

    Thanks in advance.


    Bill Thacker
    Tuesday, November 22, 2011 1:52 PM
  • I never got help for resolving this via the TechNet Forum, so I had to open a case with Microsoft Professional Support. We ended up having to implement a workaround (custom monitor) that still can't extract the free space amount in MBs although it does provide the percent free.

    Microsoft acknowledged that this was a bug/deficiency in the product (management pack), so they didn't charge me for the incident. I'm now waiting patiently for a revision to the Windows Server management pack that will address this issue.


    Bill Thacker
    • Marked as answer by Bill Thacker Friday, December 30, 2011 4:44 PM
    Friday, December 30, 2011 4:44 PM
  • Hi.  I had an issue with the Logical Disk Fress space also.  I could see when I queried the DW that a warning alert fired correctly, but a critical alert was never fired.  I spoke with MS Support and they told me that there is a problem with the monitor in all base OS management packs 6.0.6958.0.    The previous version was written so that the logical disk free space monitor triggers an alert when the state changes from healthy to critical (it was a 2 state monitor) – to match that, you would have to change the settings to alert on “critical” and not “critical or warning.” So if you have the default settin on monitor which is fire on critical or warning it only alerts once on warning and never sets to critical. 
    • Proposed as answer by Lori Getz Friday, January 6, 2012 3:27 PM
    Friday, January 6, 2012 3:26 PM
  • Bill - have you received a reply from microsoft in this respect to date?

    Thursday, March 15, 2012 10:33 AM
  • I never got an estimate with regard to a date, but I keep checking the Management Pack Catalog every week hoping it will eventually show up.

    It's disappointig that a revised Management Pack hasn't been provided after 4+ months, because having the Logical Disk Free Space Monitor work "correctly" (as it always did) is very important.


    Bill Thacker

    Thursday, March 15, 2012 12:21 PM