We have recently started using SnapManger for Exchange and since we are now doing more snapshots we need to reclaim as much space as we can on the Thick Volumes we use to keep as many SnapShots as we can.
- Data ONTAP 7.3.5P1
- FC SAN only. No NFS (NAS)
- Window Server 2008 R2 sp1
- VMware ESXi 4.1
- SnapDrive 6.5
- SnapManger for Exchange 6.04
- NetApp PowerShell Toolkit 2.3
- LUN's created through SnapDrive directly on the VM so they are RDM's
- All Volumes = Thick
- LUN's = Thin
We have an issue with changing LUN's from THICK to THIN. We created THICK LUN's originally, we then changed them to be THIN, ran both SnapDrive & Powershell versions of Space reclaim tool but the LUN's still report the THICK size on the storage (WAFL) even through Windows NTFS says the Drive/LUN is only 10% (10gb) full. We are seeing this problem on both out Production and DR sites which is across two sets of clustered Filers. This problem is occurring for all our LUN's.
Trouble shooting steps:
Used the SnapDrive Resize tool to resize the LUN down to 20GB. On WAFL it says the LUN is now 20gb but still 100% full.
I then resized the LUN to be 100gb again and when I checked WAFL, it showed the LUN was still 20GB which is great.
I then created a large 40gb file in windows, copied it to the drive which I just resized out to be 100gb.
I checked WAFL again and the LUN showed it was using 50GB in WAFL, 50% of it's capacity.
I then deleted the 40GB file so windows showed the LUN / Drive to be 10gb again.
I then ran the Powershell Toolkit version 2.3 "Invoke-NaHostSpaceReclaim" cmdlet and checked the LUN on WAFL which still reported to be 50gb even thought windows show's 10gb.
I then ran the SnapDrive 6.5 Space Reclaim tools, check WAFL and it still says 50GB even through windows shows 10Gb.
I have also changed the Volumes to be thin which did not help. The Volumes did actually give the space back to the Aggregate after becoming thin unlike our LUN's.