Comments on: LUN Clones vs. FlexClones http://blog.scottlowe.org/2007/05/21/lun-clones-vs-flexclones/ The weblog of an IT pro specializing in virtualization, storage, and servers Wed, 08 Sep 2010 12:43:11 +0000 http://wordpress.org/?v=abc hourly 1 By: Mouli http://blog.scottlowe.org/2007/05/21/lun-clones-vs-flexclones/comment-page-1/#comment-43333 Mouli Sun, 11 Jan 2009 17:16:46 +0000 http://blog.scottlowe.org/2007/05/21/lun-clones-vs-flexclones/#comment-43333 When we have 1:1 relationship between the lun : flexvol , it means i have only a single file in the filesystem. How do we create a lun clone in this scenarion. Ideally, the filesystem doesnot have any space for the lun clone ?? When we have 1:1 relationship between the lun : flexvol , it means i have only a single file in the filesystem.

How do we create a lun clone in this scenarion. Ideally, the filesystem doesnot have any space for the lun clone ??

]]> By: Jugoslav Djajic http://blog.scottlowe.org/2007/05/21/lun-clones-vs-flexclones/comment-page-1/#comment-43282 Jugoslav Djajic Wed, 07 Jan 2009 22:09:53 +0000 http://blog.scottlowe.org/2007/05/21/lun-clones-vs-flexclones/#comment-43282 Vijay, you are right, there is a 500 FlexVol limit, but unfortunately it's not per aggregate, it's per filer. Vijay, you are right, there is a 500 FlexVol limit, but unfortunately it’s not per aggregate, it’s per filer.

]]> By: Vijay Raghavan http://blog.scottlowe.org/2007/05/21/lun-clones-vs-flexclones/comment-page-1/#comment-43266 Vijay Raghavan Tue, 06 Jan 2009 22:12:17 +0000 http://blog.scottlowe.org/2007/05/21/lun-clones-vs-flexclones/#comment-43266 On the question of one lun per flexvol, isn't there a smallish limit on the total number of flexvols that an aggregate supports? On the question of one lun per flexvol, isn’t there a smallish limit on the total number of flexvols that an aggregate supports?

]]> By: slowe http://blog.scottlowe.org/2007/05/21/lun-clones-vs-flexclones/comment-page-1/#comment-35688 slowe Thu, 21 Feb 2008 00:47:42 +0000 http://blog.scottlowe.org/2007/05/21/lun-clones-vs-flexclones/#comment-35688 BA, I'd recommend you take a really close look at NetApp's SnapManager products. SnapManager for SQL and SnapManager for Oracle have hooks into the DB engines to ensure that the Snapshots are consistent and recoverable, and they provide automation tools to make the whole process easy. Otherwise, the process is certainly doable, but requires downtime from the database (you need to stop the database in order to ensure a consistent Snapshot) and some manual scripts to create clones for development and testing. Good luck! BA,

I’d recommend you take a really close look at NetApp’s SnapManager products. SnapManager for SQL and SnapManager for Oracle have hooks into the DB engines to ensure that the Snapshots are consistent and recoverable, and they provide automation tools to make the whole process easy.

Otherwise, the process is certainly doable, but requires downtime from the database (you need to stop the database in order to ensure a consistent Snapshot) and some manual scripts to create clones for development and testing.

Good luck!

]]> By: BA http://blog.scottlowe.org/2007/05/21/lun-clones-vs-flexclones/comment-page-1/#comment-35655 BA Wed, 20 Feb 2008 04:58:03 +0000 http://blog.scottlowe.org/2007/05/21/lun-clones-vs-flexclones/#comment-35655 I'm very new to ONTAP and lun/volume creation, so I am hoping to get some help :) I want to take a nightly snapshot of production database lun and present it to a development server for the devs to hammer on. What would be your recommendation for doing so, as automatically as possible of course :) Can we hammer thru this together? thx!! I’m very new to ONTAP and lun/volume creation, so I am hoping to get some help :)

I want to take a nightly snapshot of production database lun and present it to a development server for the devs to hammer on. What would be your recommendation for doing so, as automatically as possible of course :) Can we hammer thru this together? thx!!

]]> By: slowe http://blog.scottlowe.org/2007/05/21/lun-clones-vs-flexclones/comment-page-1/#comment-35158 slowe Tue, 22 Jan 2008 11:59:17 +0000 http://blog.scottlowe.org/2007/05/21/lun-clones-vs-flexclones/#comment-35158 Jeremy, Yes, you're correct in that using a 1:1 relationship between LUNs and volumes does allow for independent Snapshots of each LUN, since Snapshots run on a volume level. The ability to do a LUN clone for no additional charge, though, does make LUN clones useful in some scenarios. Thanks! Jeremy,

Yes, you’re correct in that using a 1:1 relationship between LUNs and volumes does allow for independent Snapshots of each LUN, since Snapshots run on a volume level.

The ability to do a LUN clone for no additional charge, though, does make LUN clones useful in some scenarios.

Thanks!

]]> By: Jeremy http://blog.scottlowe.org/2007/05/21/lun-clones-vs-flexclones/comment-page-1/#comment-35089 Jeremy Wed, 16 Jan 2008 16:33:26 +0000 http://blog.scottlowe.org/2007/05/21/lun-clones-vs-flexclones/#comment-35089 That's one of the reasons that the best practice is to have a 1:1 lun:vol relationship. Creating multiple flexible volumes doesn't take any more space apart from the snapshot reserve. Snapshot per volume, lun per volume = snapshot per lun. At least, that's my understanding. That’s one of the reasons that the best practice is to have a 1:1 lun:vol relationship. Creating multiple flexible volumes doesn’t take any more space apart from the snapshot reserve. Snapshot per volume, lun per volume = snapshot per lun.

At least, that’s my understanding.

]]>