a Question !
I have an EMC Networker data protecton suite and two devices in my scenario ,
1- Data Domain
2- Tape Library
I want to run my backup plan in this way, that :
- in first step backups of my data write on data domain (with networker) . the retention time of this backups is (about) 1month. (Backup action)
- in second step I want to write my backups on Tapes. and its retention time is about 1Year. (Clone Action)
in my scenario, I wont to write data in two action (in workflow) that DONT write in same time !
I need that writing data on my tapes, will start when the data on data domain are becoming expired!
I mean that in this scenario,
in the first day, backup action just run (and clone job do not run) - this data protected on data domain 30 days.
in 29th day, I want to run Clone action, that transfer data from data domain to Tapes ! (and then it protect for 1Yesr on tape)
is there any solution on EMC Networker for do this job ?
You don't say which version of NW you run (as retention policies changed in NW9), but we regardless I will assume here NW9 methodology as soon or later you will hit it even if you are running NW8 now. Given that b/r can't be difference there, you would not clone the record, but rather stage it (which is also handled by nsrclone if using script). Or you will have to define staging policy which takes records from DD after 29 days and moves them to tape pool.
We do it in a similar way. However, we clone right after the save sets have been backed up to the DD.
As the disk save sets will expire separately, they will do that independently, earlier.
However, this method gives you more time to react in case of a failure.
Think about it:
If you do clone/migrate at the end of a cycle and you run into whatever problem, it might be that NW would have removed the backup from disk until you finally manged to continue.
BTW - we use scripted cloning for this procedure.
hi Hrvoje Crvelin
the version of NW that I`m using it ,is 9.0.1.x
but , what`s the difference between definition of staging and Policy ?
I thought about you said, your solution is Logical and more safe,
but I want to know is there any way to store backups on separate volumes ? in this case, backups store for 30 days on DD and when it neared to expiration on DD, write on Tape with about 1 year retention.
dear Hrvoje Crvelin
I`m thinking about your solution, "Staging" .
I should test it and if was my wish, I`ll tell you about this.
but, if where is any other solution, please tell me ...
"Is there any way to store backups on separate volumes ? "
This is the usual setup. You configure 2 pools:
- 1 backup pool (DD) with retention of 1 month
- 1 clone pool (tape) with retention of 12 months/1 year
As the save sets expire separately, you do not have to do anything else.
And yes - of course you can recover from a clone. Otherwise it would not make sense.
Difference between cloning and staging is that one copies the data and second one moves the data. I do not have NW9 to check exact steps there now, but in theory since you wish to extend existing backup at the end of expiration period to me it sounds as you wish to extend retention once data is moved to different media.
according to emc documents (duco61081) , (in my pilot environment) I create a Staging Device. and I refered it to one of my Devices (a AFTD device) and I set that he oldest save sets move to a new media pool (that is DD) .
and then, I configured a policy (and workflow) to get backup on a media pool that refers to AFTD device.
how staging works?
did it moves last save sets automatically to new destination pool (that I defined it DD) ?
I mean that when we define a Clone action, we define a new dedicate definition on workflow.
But how should define a Staging to action?
Staging was necessary in the legacy times (when disk capacities were much smaller) - today it is almost history.
Especially staging from an AFTD to a DD does not really make sense as you will loose a lot of features like 'client direct' (to a DDboost device) which will really speed up you backup throughput.
In general staging is controlled by the 2 watermarks:
- high watermark is where the staging should start
- low watermark is where it should end
The most important issue: the % values refer to the whole file system.
Other parameters are of 'minor' importance like timing, sequence etc.