My appologies if this has been addressed before.
I have an NFS volume on my FAS2240 that does not reclaim disk space when I move a VM off via Storage vMotion. All other volumes do.
Move 40GB VM from Saphire_Datastore to Gold_Datastore.
Volume free space prior to move:
Saphire - 108GB Gold - 328GB
Volume free space after move:
Saphire - 108GB Gold 290GB
I am relatively new to both VMware and NetApp storage, so I am unsure what direction to take. I have looked online, but most of the leads I have located relate to deleting VMs or reclaiming snapshot space.
Update: I just noticed in OnCommand System Manager that there are a group of tabs at the bottom of the Volume pane. The Saphire volume has 423GB of Snapshot data even though I have left the snapshot space at the default 5%. I guess it is safe to delete the larger snapshots, but I wonder why they would be so big in the first place. One snapshot is 256GB at was taken at 4:00 p.m. yesterday. It just happens that I was migrating a 250GB VM from Saphire to another datastore at about that time. Has anyone seen this type of behavior?
I believe I have figured out my issue: My snapshots were configured to run at the default times. One of those times is 4:00 p.m. The snapshot is so huge because the file was still in the process of being copied. Well, now I have a good reason to change those times!
you don't want snapshots just go to the volume resize and
Snapshot reserve (%):00,
Then it will show, total allocated space as volume space.
Snapshot is working on copy on first write technology,so what ever may be updates happen it will take as snap that's why when you migrate the 250GB VM its taken entire snap.
Just want to explain this a bit more,
Its true you should put on NFS and CIFS snap reserve at 0% since higher number will cause you to save space for snapshots that can be use to data
vol size 100mb
sharing with nfs or CIFS will give you 100mb
if you give snap reserve 5 this will make the system save 5% for snapshots not matter what ... meaning
users will see 95mb
if you put 0% users will see 100mb and you can still use snapshots (it will take space for the 100mb)
if you use 5% and snapshots take more then 5% it will take space from the remaing 95mb
as you can see snap reserve 0 make sense
just a side note : its not like this in LUN's
hope this explain abit more
Was your question answered correctly? If so, please remember to mark your question Answered when you get the correct answer and award points to the person providing the answer. This helps others searching for a similar issue.
Was this ever answered?
We have about 50GB of files that were deleted from a VOLUME over a month ago. THe problem is that none of the space seems to have been reclaimed. This is a netapp 250 running ontap 7.2.xx. It is used as an nfs server for a couple of Solaris boxes in out network
yes they do use snapshots. Weird thing is when on the solaris hosts I do a du-h command on the .snapshop dir and it seems like it's as big (168GB) as the volume itself
/vol/vMedia/ 200GB 170GB 29GB 85% /vol/vMedia/
/vol/vMedia/.snapshot 50GB 5705MB 44GB 11% /vol/vMedia/.snapshot
netapp2> snap list
%/used %/total date name
---------- ---------- ------------ --------
0% ( 0%) 0% ( 0%) Apr 04 09:26 netapp1(0084329016)_vMedia.1266 (snapmirror)
0% ( 0%) 0% ( 0%) Apr 04 08:00 hourly.0
1% ( 0%) 0% ( 0%) Apr 04 00:00 nightly.0
1% ( 0%) 0% ( 0%) Apr 03 20:00 hourly.1
1% ( 0%) 1% ( 0%) Apr 03 16:00 hourly.2
1% ( 0%) 1% ( 0%) Apr 03 12:00 hourly.3
1% ( 0%) 1% ( 0%) Apr 03 08:00 hourly.4
2% ( 0%) 1% ( 0%) Apr 03 00:00 nightly.1
2% ( 0%) 1% ( 0%) Apr 02 20:00 hourly.5
3% ( 2%) 2% ( 1%) Mar 31 00:00 weekly.0
Just thought id add my experience in case anyone comes looking here to reclaim deleted blocks from the Windows O.S. in NFS mounts presented to VMWare. I was looking and researching for a while until I found Netapp VSC.
With the Netapp VSC plugin installed on VMWare virtual centre, this enables a reclaim space feature to VMWare. Running this process would power down the VM, reclaim the deleted logical blocks from the Windows OS, freeing them up on the storage array. When it automatically powers the VM back on the VM Used consumed space was decreased / true used.