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.

Remove stale/obsolete volumes and applications from SAM?

We were running low on available monitored volumes, so I started attempting a cleanup to get rid of some of the stuff that we don't care about that snuck in through laziness or inattention when setting up nodes in the first place, or just for things that have been reconfigured.

What I'm finding, though, is that monitored volumes that no longer exist on the target node can't be removed.  That they show up green even though they're not present is a whole other issue...

E.g. this server had an iSCSI disk mounted as Z: and sometime in the distant past, someone chose to monitor it.  That iSCSI device hasn't been around in over a year.  Trying to remove it from monitoring is not doable, since it's not shown in the Resources tree.

SAMstaleVolumes.PNG

SAMstaleVolumes2.PNG

I've tried using the "None" option, hoping it would strip all monitored resources, but I was left with JUST the obsolete Z: drive. emoticons_happy.png

Short of deleting all nodes that exhibit this sort of thing and re-adding them, how do we work around this?

Same thing for a couple of servers that at one time were running Hyper-V (they never should've had the role installed in the first place, but I digress) and have since had the HV role removed.  They still show up under the Virtualization section with red icons.  I can't uncheck Hyper-V on them, since it doesn't show up.

  • "List Resources" shows only what's available to be monitored, not what is monitored but no longer available. To remove obsolete volumes or interfaces from monitoring you can do so using "Manage Nodes" located under "Settings" .

    Delete Volumes.png

  • Nice, thanks.  Never even noticed that Delete button way over there.  I've only ever added or removed stuff from the List Resources page.

    That's letting me kill the stale volumes.

    Any idea how to fix the similar issue with former Hyper-V servers?  I can disable polling, but not remove it.

  • This works and is the only way I know to get rid of the evil grey interfaces or volumes...

  • [Settings -> Virtualization Settings -> Hyper-V Settings]

    Virtualization Polling Settings.png

  • Right, like I said, I can disable it, but is there a way to get rid of it entirely?  I'd like to de-clutter the tree, but disabling polling leaves those servers in there, just gray.

  • Ideally Hyper-V would have been disabled in Orion before it was removed from the host. If it was, then you would have been able to easily disable this through "List Resources". Unfortunately now the only way to disable Hyper-V is directly through the database, similar to Hardware Health.

  • You can try removing HyperV hosts with disabled polling manually in database using following SQL statement:

    DELETE FROM [VIM_Hosts] WHERE

    ManagedStatus = 0 -- Polling is disabled

    AND PlatformID = 2 -- HyperV

    After this you may want to run database maintenance to remove orphaned child VMs of all removed hosts.

    Note that it is strongly recommended to make backup of database before doing any changes manually.

  • Thought I did this months ago, but apparently I got distracted by something shiny...

    That did the trick, though, thanks!

  • Hello, I checked in our environment and found there are over 3000 stale interfaces/volumes/hardware sensors. Is there any Ideal and easy way to search -select and delete these through SQL commands ?

    or going through one by one and deleting them from the UI is the only option?

    Thanks in Advance,

    Chetz

  • This is the select I used to find 684 rows that have not sync’d in the last 11 days. You can see a couple of them say volume is responding.

    select [VolumeID],[fullname],[VolumeDescription],[VolumeType],[StatusLED],[VolumeResponding],LastSync,[Volumes_EventID],[VolumeWarningLevel],[VolumeCriticalLevel]

       FROM [dbo].[Volumes]

       where [volumetype] = 'fixed disk' and LastSync < dateadd(dd,-11,getdate())

       order by LastSync

       ______________________

    This one shows 2986 rows where volume is not responding. 

    select [VolumeID],[fullname],[VolumeDescription],[VolumeType],[StatusLED],[VolumeResponding],LastSync,[Volumes_EventID],[VolumeWarningLevel],[VolumeCriticalLevel]

       FROM [dbo].[Volumes]

       where [volumetype] = 'fixed disk' and VolumeResponding = 'n'

       order by LastSync

    I then exported results to an excel to validate. I also see coordinating events in message center that shows "volume removed", etc so I am  trying to create an alert to auto fix.

    Could we resolve the issue dynamically but using network discovery of nodes?