Hi
Please take a look at TR 3737 . This is the best practices guide for SMVI 1.0. But this should get you started on some of the best practices for retention, scheduling etc.
http://www.netapp.com/us/library/technical-reports/tr-3737.html
We are currently working on 2.0 best practices and I'll let you when that's done.
Regards Amrita
-----Original Message----- From: Raj Patel [mailto:phigmov@gmail.com] Sent: Friday, November 13, 2009 2:40 AM To: toasters@mathworks.com Subject: SMVI Experiances
Hi all,
Now our controller upgrade woes have been sorted I can start to play with the new stuff on our 2050.
One of those is SMVI.
Unfortunately as we're not on vSphere (and our ESX hosts aren't up to 3.5 update 4) we'll be sticking with v 1.2 of SMVI until we upgrade our ESX infrastructure to v4.
Anyone out there willing to share their experiances with SMVI ? It looks fairly simple to configure and get going but just as I was about to schedule a few backup jobs I thought I'd trawl the net to find out if there are any 'gotchas'.
For example we use iSCSI - will the VM grind to a halt (ie queisce) while the Volume is snapped (ie is there a performance hit) or is this a momentary operation (ie safe to do during the day) ?
I have seen the note about snapping a VM with iSCSI attached LUN's - I can live with that (not sure about the DBA's . . .).
For Prod we tend to put a 300Gb VMFS lun in a 500Gb flexvol - rate of change is generally pretty minimal for the app servers (generally w2k3 or w2k8 servers from templates). Is this a workable ratio or do I need to allocate more space (I'm looking at just a once a day snap with a 1 or 2 day retention) ?
Thanks in advance, Raj.