Amanda-Users

Re: Overwhelmed...

2005-10-07 11:01:31
Subject: Re: Overwhelmed...
From: Joshua Baker-LePain <jlb17 AT duke DOT edu>
To: CG <cgg007 AT yahoo DOT com>
Date: Fri, 7 Oct 2005 10:48:39 -0400 (EDT)
On Fri, 7 Oct 2005 at 7:07am, CG wrote

I'm a bit overwhelmed by the daunting task of having to develop a backup schema
for my network, then create configuration files to match. I've read manpages,
and FAQ's and docs. The documentation is extensive, and the learning curve
looks really steep from here. Is there a quickstart guide? Some case-studies? A
good place for me to start to wrap my brain around the AMANDA concept of
backup?

The chapter "Installing AMANDA" <http://www.amanda.org/docs/using.html#installing> in the docs is a pretty decent quickstart.

The main thing to wrap your head around is that amanda is basically a (very good) backup scheduler. What actually gets the bits off the disks is dump or tar (yes, run by amanda). But amanda's main task is scheduling, and it does it differently from the "old" *nix paradigm of "full on friday, incrementals the rest of the week". Amanda's goal is to backup about the same amount of data each night by spreading the full backups throughout the dumpcycle. It does this within the 3 main parameters you specify regarding the schedule:

dumpcycle = max amount of time between level 0 dumps of any DLE (disklist entry)
runspercycle = # of times you'll run 'amdump' per dumpcycle
tapecycle = # of tapes you'll give amanda to use, generally recommended to be >= 1.5*runspercycle

--
Joshua Baker-LePain
Department of Biomedical Engineering
Duke University

<Prev in Thread] Current Thread [Next in Thread>