Virtual Tape Library Data Move

Networker does not stage from tape to tape

People I know with such configuration do a clone with greater retention to a tape. In your case, clones would have 5 weeks retention and original tape (EDL) would expire after 2 weeks, so you can re-use this (and in case of restore, use the clone ones).

Wow, I always believed that was not possible. As the manual says:

"A save set can be staged from one disk to another as many times as required.
For example a save set could staged from disk 1, to disk 2, to disk 3, and finally
to a remote tape device. Once the save set is staged to a tape device, it cannot
be staged again."

What I am missing here?

Daniel, you can stage from tape to tape. You will need different pools (never tried same pools so that might not work) and that's all. Of course, as original question goes, you can do that with scripted staging.

You are misinterpreting last sentence (actually, that sentence is written badly). Last sentence doesn't say you can stage further from tape to tape, but rather once you staged from disk to tape you can stage it again from disk to another tape as staging is moving and not coping. I agree that the way as it says is misleading.

I used tape2tape staging on several occasions when eg. daily backup had to be changed to monthend so first you would change policy of saveset and then you would stage it from one tape pool to another (as pools were policy oriented).

I am in the same position. I backup to a VTL and then immediately clone to a set of "clone" tapes that are taken off-site. This means I have a copy on-site on the VTL and a copy off-site on the "clone" tapes. I have written scripts which use nsrclone to copy savesets and nsrmm -d to manually delete savesets from the VTL which emulate the behaviour of nsrstage on an AFT device.

It is not pretty but it works.

Hi Hrvoje ,

My Question is can we not do the staging by using the Auto staging just only defining the lower and upper water level percentage


Thanks

Automatic staging works for disk. If you wish to do it from tape to tape you must use nsrstage command manually.

Hi,

Since when does nsrstage work for tape to tape staging? Is it a new feature in 7.4? I have attempted with 7.3.3 and it just gives an error!

Rob

Since I work with NW at least. Last time I used it it was with 7.2.2. You must make sure to use nsrstage command and ssid/cloneid pair and it will work.

Hi Hrvoje,

Manual staging from tape to tape works!!! but it throws an error at the end. I have just tested manually staging from VTL tape to physical tape. It successfully cloned the saveset, removed the original saveset from the source tape but then displayed an error message "nsrstage: Recovering space failed with the error Space can only be recovered from adv_file and file type devices."

I always assumed this error indicated that it could not delete the saveset instance from the source tape. Do you know how to tell it not to attempt to reclaim space? There's nothing in the man page. See the verbose output below:

root@legolas# nsrstage -v -m -b Daily -S 824372240/1210248208
Obtaining media database information on server legolas
Parsing save set id(s)
Migrating the following save sets (ids):
824372240
Automatically copying save sets(s) to other volume(s)

Starting migration operation...
...from storage node: legolas
nsrstage: Successfully cloned all requested save sets
nsrstage: Clones were written to the following volume(s):
E02422
nsrstage: Recovering space failed with the error Space can only be recovered from adv_file and file type devices..

That's ok. With disk devices you can remove saveset from original source, with tape devices you can't. That's the nature of tape standard.