Meanwhile i read the documents and it looks like ndmpcopy is the best option as i have limited downtime (3hrs total). I will also try out VFM and robocopy on the simulator.

I would like to thank each one of you for taking time to answer the queries. We have ordered few more Filers and i hope to improve my knowledge of the netapp systems by then. If I get stuck somewhere, I now know whom to ask... :)

Thank you very much.

On 8/22/07, Chris Dittert cdittert@brocade.com wrote:

Lohit,

Take a look at NetApp's Virtual File Manager – Migration Edition (VFM-ME), it will address your needs.

•      Migrates data from source share/folder to destination
  

share/folder

•      Uses automated policies to set up migration tasks including
  

creation of shares on destination, performing baseline migration copy, bandwidth throttling, establishing incremental copy schedule, hiding the source share to perform final copy & selecting whether to retain ACL information or not

•      Retains all file attributes including date/time stamps to
  

provide accuracy of current and/or future ILM data classification

•      Has ability to setup & manage MSFT DFS to make this migration
  

transparent to users (if you are using or want to implement DFS) by automatically updating DFS links to new share/folder targets when data migration is complete = zero downtime

http://www.netapp.com/products/enterprise-software/manageability-software/da...

*Chris Dittert*

*From:* owner-toasters@mathworks.com [mailto:owner-toasters@mathworks.com] *On Behalf Of *Goodin, Jeff *Sent:* Wednesday, August 22, 2007 8:26 AM *To:* toasters@mathworks.com *Subject:* RE: copying qtree

Depending on the rate of change, file size and number of files in this Qtree, robocopy may work well. We generally use it to move large amounts of windows data between volumes/filers when we can't use snapmirror. It will keep the ACL information for you if you give it the right flags.

The first copy will take a while, but you can do file level incremental updates after that.

You can download it from Microsoft. It is part of the Windows Server 2003 resource kit. The kit includes a robocopy.doc word document that explains all of the flags.

At the time of the cut over, delete the share (after saving the share settings), do a final robocopy and recreate the share pointing to the new location.

Jeff.

---

*From:* owner-toasters@mathworks.com [mailto:owner-toasters@mathworks.com] *On Behalf Of *Holland, William L *Sent:* Wednesday, August 22, 2007 6:18 AM *To:* Lohit; Fox, Adam *Cc:* toasters@mathworks.com *Subject:* RE: copying qtree

You can use ndmpcopy or vol copy to do this.

---

*From:* owner-toasters@mathworks.com [mailto:owner-toasters@mathworks.com] *On Behalf Of *Lohit *Sent:* Tuesday, August 21, 2007 11:41 PM *To:* Fox, Adam *Cc:* toasters@mathworks.com *Subject:* Re: copying qtree

Sorry for the late response, i live in a different timezone. We don't have snapmirror license :(

On 8/22/07, *Fox, Adam* < Adam.Fox@netapp.com> wrote:

If you have a SnapMirror license, I would do a qtree snapmirror. Set up an incremental schedule, say once an hour.

Then when you're ready to cut the clients over, you'd only move an hour's worth of changes over which should be

quick.

-- Adam Fox adamfox@netapp.com

---

*From:* Lohit [mailto:lohit.b@gmail.com] *Sent:* Tuesday, August 21, 2007 3:21 PM *To:* toasters@mathworks.com *Subject:* copying qtree

Hello experts,

I am very new to the netapp world. Sorry if this has been answered already or if it sounds stupid.I have the following situation,

What we have is a FAS940 with DS14MK2 shelves(3nos).

There is a qtree "proj-CFD" which is 200GB in size and the containing volume is 500GB. The volume is 90% full. I have been asked to increase the space for the qtree to 400GB. Previous admin created the aggregate(Aggr1) with a raidsize of 8 and there are two raidgroups with 8disks in each raidgroup.

I have three spares and adding these to the current aggregate would create another raidgroup of 3 disks and i will be left with 1 usable disk (we have raid_dp) and also i will have no spares left.

But fortunately there is another aggregate(Aggr2) with a raidsize of 16 and there are 14 disks in that aggregate now. Now my question is, how can move the "proj-CFD" qtree to a volume in Aggr2 with minimum downtime and keeping all the ACLs intact(It is ntfs only).

Thanks

-- With warm regards

Lohit

-- With warm regards

Lohit