Networker

Re: [Networker] Stage savesets from AFTD to tape

2009-11-11 16:14:46
Subject: Re: [Networker] Stage savesets from AFTD to tape
From: Tom Birkenbach <tom.birkenbach AT WMICH DOT EDU>
To: NETWORKER AT LISTSERV.TEMPLE DOT EDU
Date: Wed, 11 Nov 2009 16:12:38 -0500
Good point!  Thanks Preston!

First, for the initial backup, what I'm doing is saving to AFTD and then
cloning to tape as per the setting in the group.  What I'd like to do is
stage my original savesets from AFTD (for performance reasons and so as to
not have to read from the clone tapes) to tape.  I should then have two
copies on tape, but in two different tape pools.

So what you're suggesting is that if done incorrectly, all references to my
original clone would disappear when I attempt to stage from AFTD to tape, right?

On Thu, 12 Nov 2009 07:48:36 +1100, Preston de Guise
<enterprise.backup AT GMAIL DOT COM> wrote:

>Hi Tom,
>
>On 12/11/2009, at 07:20 , Tom Birkenbach wrote:
>
>> In an effort to stage savesets from AFTDs to tape, I'm using the
following procedure.
>> 
>> #mminfo -ot -q 'location=storagenode1,copies=2' -r ssid > stagelist.txt
>> #nsrstage -v -m -d -b LTO3Pool -S -f ./stagelist.txt
>> 
>> I believe my only real "gotcha" is if I forget to set the location on the
>> AFTD.  Ultimately, what I'm trying to do is stage all the savesets from the
>> AFTD to tape to clear the disk (for maintenance purposes, disk resizing,
>> whatever...).  My method feels a bit "hackey" and, with the multitude of
>> Networker commands, it seems like there could be a better way.  Any ideas or
>> suggestions?
>
>The absolute number one suggestion I have for you is DON'T specify just the
saveset id.
>
>(Sorry for the caps but it's that important.)
>
>If you're cloning out to tape first, then subsequently stage, then staging
specifying just the saveset ID will have unintended consequences.
>
>If you stage by saveset ID alone, NetWorker will delete _all_ details of
_all_ instances of the saveset except the instance you just staged out.
I.e., you lose reference to your clones as well.
>
>Always run mminfo for staging operations as:
>
>mminfo <rest> -r ssid,cloneid -xc\/ | grep -v ssid
>
>That way, mminfo will construct the results including saveset ID and clone ID.
>
>Also, because of that, you'd want to change your query field from
'location=X' to say:
>
>volume=Y.RO
>
>where Y.RO is the read-only volume you want to stage from; run once per
disk backup unit.
>
>I realise with you using a query of "copies=2" that you're probably not
cloning --- however, honestly, staging by just saveset ID alone is something
you want to avoid getting in the habit of; it's just too dangerous.
>
>Cheers,
>
>Preston.

To sign off this list, send email to listserv AT listserv.temple DOT edu and 
type "signoff networker" in the body of the email. Please write to 
networker-request AT listserv.temple DOT edu if you have any problems with this 
list. You can access the archives at 
http://listserv.temple.edu/archives/networker.html or
via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER