2014-03-13 05:08 AM - edited 2015-12-18 12:25 AM
I am new to Snap Creator so please pardon may ignorance.
I am looking to take crash-consistent snapshots for multiple volumes across different controllers which may be assigned to ESX nodes or VM. Does NetApp snapcreator has the capability to create consistency groups which may span across separate controller and take crash-consistent snapshots on them. If yes, what are the supported platforms?
2014-03-13 05:12 AM
SnapCreator will take CG snapshots across as many controllers as you configure. It sets up the write fencing one by one, which might seem odd at first unless you look very closely at how it works. The end result is that all dependent write order is preserved. I’ve personally done snapshots across up to 6 controllers as far back as 6 years ago. Never had a problem…
2014-03-13 05:17 AM
You can take consistency group snapshots with SC but the consistency groups are per controller, so you get one for each controller. It is a two phased process 1) cg start 2) cg commit.
SC uses following process assuming controller1:vol1,vol2 controller2:vol3,vol4
1) cg start controller1:vol1,vol2
2) cg start controller2: vol1, vol2
3) cg commit controller1:vol1,vol2
4) cg commit controller2:vol3,vol4
While all volumes are fenced via the cg start everything is consistent. However the commit where fencing is removed and snapshot taken is not guaranteed to happen at same time for both controllers and thus consistency cannot be guaranteed unless the application using the storage issues write dependent writes. Most if not all databases do this for example. This means if first write is not acked then another write wont be sent. If this is the case the combination of that plus the CG mechanisms ONTAP provides guarantees consistency. If this isnt the case you need to get application consistency.
Unfortunately I cant give you a yes or a no but hopefully this helps you drive toward a solution
2014-03-13 05:22 AM
What’s the application? The only app I know where CG snapshot might not work is Teradata and it’s possible to configure reiserfs in a really dangerous way. For mainstream filesystems and applications, no problem. If the SC operation completed, your snapshots are valid.
2014-03-13 06:58 AM
2014-03-13 07:18 AM
Thanks Keith and Steiner.. guess I will have to test this to know better.
This is to make sure we have point in time backup copy available of the Delphix dSource in case of logical physical corruption. Because if there is corruption on dSource, you basically loose all your vDBs.