Snapview session and snapshot

Please consider moving this question as-is (no need to recreate) to the proper forum for maximum visibility.  Questions written to the users' own "Discussions" space don't get the same amount of attention and can go unanswered for a long time. 

You can do so by selecting "Move" under ACTIONS along the upper-right.  Then search for and select: "VNX Support Forum".

Sampath_Acc wrote: 1. Can we able to restore the source LUN using Snapview session created on it . ( No separate snapshot was created and associated to the session )

You are absolutely correct.  You do not have to first activate the SnapView session via a snapshot (well... technically you activate a snapshot which makes it available to a session) to restore the production LUN with the data as represented by the session (your point-in-time "view").

2. Can we able to restore the source LUN using the snapview snapshot only . I can understand snapshot can be presented to the storage groups for testing and backups ... Any suggestions would be appreciated.

If in your context above restoring the source LUN means overwriting it with the data as represented by the session (since it is pointer based really just the changes saved to the RLP need to be restored), then as mentioned above, no you don't have to present it first via a snapshot.  However, it may make sense to if for instance you want to first verify the contents or for instance only pull out some of the data since the restore within Unisphere is all-or-none.

I'm thinking what may help is the analogy of CD's and a CD Player.  Think of the SnapView sessions as the CD's which of course contain the data/music and represent the point-in-time (view).  Within Unisphere you are creating your CD's when you start the session.  Now think of the snapshot as the CD Player used to playback the session/CD.  The snapshot or CD Player doesn't by itself have any data, but is used to playback the data/music; the snapshot is merely a container required to present it to a host.  One more thing, just like you don't need a CD Player for each CD you own, similarly unless you want to listen to more than one CD (or present more than one session) at the same time you can reuse the same snapshot.

Does this help a bit?

Also, you may have already reviewed them, but since you own a VNX and if running VNX OE for Block v32, there is another option for your pool LUNs only which is the VNX (Advanced) Snapshots.  Main differences are that it doesn't require a separate explicit allocation of LUNs (RLP) but instead uses the remaining space in the same pool, instead of CoFW for the pointer based implementation it uses RoW (redirect-on-write), and it is worth mentioning that there is limited support for VNX Snapshots even with our own layered applications currently.  However, from what I can tell sounds like you are managing the feature manually via Unisphere (maybe Navisphere CLI/admsnap) and not automating them via Replication Manager for instance which is one example that doesn't support VNX Snapshots.

Thanks Chris,

Discussion moved to right forum now. We have VNX snapshot in our environment. We are not using it as it occupies space from the same pool where the source LUN resides. Normally we use raid groups for RLP.

In NetApp we can set the snapshot space reservation to limit the utilization in the source volume.

Do we have any space reservation set by default in VNX for VNX snapshots? if so how much, and where i can see. If we can set manually then how to set ? Just worried that the pool will be occupied entirely if source LUN(s) having more block level changes.

I am searching for the document which explains step by step restoration of snapview session/snapshot. If you (and anyone) have the URL please post it here ...

In NetApp, you typically do not use snap reservations for volumes with LUNs because they don't do you much good.  That feature was built for file because when users delete data, they like to see free space come back and without a snap reserve that doesn't happen as long as there is at least one snapshot.  But in a block environment, the volume space is never exposed to the user since the user only sees the space inside the LUN itself so the recommendation is to not use it because at the end of the day, it's a one-way wall since snapshots can always bleed into user space if they need to anyway.

That being said, we don't use snaps reserves with VNX snapshots.  We consume extra space in the underlying pool as needed.  We do, however, have policies you can set to start deleting snapshots when the pool gets to a given utilization level.  This will work to free up space so you don't run out.  NetApp has something similar (see 'snap autodelete') which they recommend running because if their volume runs out of space, you have the same problem.