Networker

Re: [Networker] Advice needed on staging policy values

2009-09-08 17:45:05
Subject: Re: [Networker] Advice needed on staging policy values
From: Len Philpot <Len.Philpot AT CLECO DOT COM>
To: NETWORKER AT LISTSERV.TEMPLE DOT EDU
Date: Tue, 8 Sep 2009 16:40:40 -0500
> Preston de Guise 
> 
> Any change you make to the staging policy will come into effect 
> immediately, so if you want for the time being you can push your max 
> storage period out to a larger value, and then tweak it back down 
> later when you've worked out what your optimum options are.

OK - That makes sense. I saw in the admin guide where it said that edits 
to the policy take place immediately, but I wasn't sure if that meant 
"right now, from this point forward" in terms of savesets or globally, 
'past and present'. So it's global and instantaneous.

 
> As to tips/traps, the key thing of course is managing capacity of each 
> individual disk backup unit; NetWorker doesn't move backups from one 
> DBU to another if the first fills, so if that happens you'll have an 
> immediate staging requirement, which is where watermarks will come in 
> - indeed, watermarks should come in before that, but it doesn't quite 
> always happen depending on availability of tape, levels of activities 
> on the volumes and sizes of the largest savesets.

Does the high watermark apply to AFT volumes individually, or to the pool 
they belong to as a whole? For example, if you have a pool with five 
volumes and an 80% threshold, would it stage upon the first volume 
reaching 80% capacity or when the aggregate total of all five (i.e., the 
pool) reaches 80%?

 
> Best thing is if 7 day retention is too short, push it out, and 
> monitor how things go closely with a view to pulling it back down to a 
> "middle" point if a month turns out to be too long.
> 
> (As much as anything, determining optimum time on disk boils down to a 
> variety of factors including frequency of full backups, maximum size 
> of any individual saveset, number of tape drives available for cloning/ 
> staging, whether you're cloning as well, etc.)

Fortunately, in this case it's a system that backs itself up full every 
night (i.e., reasonably simple environment for the moment). Of course, 
we're gonna add some other diff/full clients to the mix RSN... :-)

Thanks for the help.

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