This discussion has been locked. The information referenced herein may be inaccurate due to age, software updates, or external references.
You can no longer post new replies to this discussion. If you have a similar question you can start a new discussion in this forum.

Mount point capacity monitoring stopped

Just opened a support case,1343324, as I had found after the upgrade to NPM 12.2 and SAM 6.4 our mount point drives on a servers are no longer polling capacity.   I learned this morning this is a recently discovered bug and I shouldn't upgrade my production system if I can help it.  The problem is the system is already upgraded.   This was not listed in the known issues on the release notes.  I understand development is working on a fix, however, just a bit of feedback here, it would be nice as these are discovered that release notes in the customer portal are updated to list out things like this.  It may be that in most environments mount points may not be a critical drive, however, in our environment, our SQL servers utilize mount points for their data and log drives.  This leaves us somewhat exposed as we now can no longer be pro-active on these drives in regards to capacity.   We will have little to know warning if these drives fill up.  Had this been in the release notes, I would have held off upgrading until a fix was released.  

  • mharvey​, we are tracking this issue internally under case CORE-9517 and hope to have a resolution soon.

  • Thanks aLTeReGo​.  I appreciate the reply.  In the meantime I'm having to build out WMI monitors for this.  Not the most fun use of my time, but hey, it's a workaround at least.  I do have a dev environment with the same issue so I am willing to test out any possible fixes that the developers come up with.  I'm none too concerned about our development deployment.

  • Same issue for us - been waiting over 30 days now for a fix :/ Hoping it will be released soon. Server admins are not too happy.

  • My company is having the same issue. 

    I opened up Case # 1325147 on 12 Oct 2017.  Same issue.  We have been using this script as a workaround https://thwack.solarwinds.com/docs/DOC-174800

    We also now have a script that looks for volumes that are not monitored and alerts on that.

  • Thanks for the heads up.  I just finished creating SAM components to monitor these via WQL and Windows Performance counters, however, it's definitely not ideal.  

  • Yeah - I am forced to currently use a seperate temporary NPM 12.1 installation to monitor the affected systems with mount points.

    Must say I am incredibly disappointed in the time it is taking Solarwinds to fix a core issue like this. Cannot understand how it can possibly take long over a month to find a fix - with no time estimates or feedback on progress :/ Disappointing.

  • This has hit us too, any updates? This is critical monitoring for our environment.

    I'll investigate the workarounds that the others have mentioned in this thread. Any other options?

  • Other that work arounds, not that I've heard of.  

    The workaround I've had to do for each of our  mounts is to create a WMI monitor of

    Capacity using:

    Select Capacity From Win32_Volume WHERE Caption = 'Drive:\\path\\'

    Freespace using:

    SELECT FreeSpace FROM Win32_Volume WHERE Caption = 'Drive:\\path\\'

    and percent free space using

    SELECT PercentFreeSpace FROM Win32_PerfFormattedData_PerfDisk_LogicalDisk WHERE Name='Drive:\\path'

    for each mount point on the server.   It takes some time and we've had tocreate new alerts for these, but it is getting us something for now.   NOt sure of the additional overhead this is causing in addtion to all the other monitoring.  And in case anyone is wondering, HF1 for the Orion core does not resolve this. 

  • The script/template mentioned by wesche below works well for a backup until this issue gets resolved. Last update we had was that they were still working on a fix for a later release. No timeline, only a promise to be updated soon.

    Only issue with the script is unless you fine tune it for each node, you'll see every volume on the device. Which can get crowded if you have SQL clusters set up with tons of mount points (not my idea).

  • Thank you very much mharvey Guess this will be my afternoon / evening. We upgraded on Wednesday and I found out about this bug this morning when I received an email from the DBAs that the Orion server SQLLogs mount volume nearly filled up. Fun morning.