[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Backup schecduling



Well, I like HD mirroring because if the main system dies, I just stick the
backup HD in and reboot, and the system is running again (either on single
user doing some forensics, or 100% functional)
I hadn't considered the option of one of my backup disks dying, still I
don't favour the idea of having just a few days to realize there's a rootkit
before it spreads to the backups... I know, 5 minutes should be enough to
detect rootkits, but just in case...
Thanks for the suggestion!
Miquel


On Thu, May 17, 2001 at 05:39:15PM -0700, Alvin Oga wrote:
> 
> hi ya miquel
> 
> assuming you want to use two hard disk backups... ( must be important data
> ) ??
> 
> main# mkdir /Backup_1
> main# mkdir /Backup_2
> 
> -- you should alternate between the two backup disks/servers
> 	- if your main disk and your backup_1 dies...you still have
> 	yesterday on Backup_2
> 
> have fun
> alvin
> http://www.Linux-Backup.net .. sample scripts ...
> 
> Use "date"... figure out if it even or odd days...
> and mount the even or odd days backup_1 or backup_2
> 
> 	miquel_backup.pl
> 	...
> 	if ( $odd ) {
> 	  mount Backup_1:/Backup_1 /Backup
> 	} else {
> 	  mount Backup_2:/Backup_2 /Backup
> 	}
> 	...
> 	--
> 	-- backup only changes since last full backup
> 	--
> 	find /home /etc /root -mtime -$cnt | tar zcvf /Backup/$DATE.tgz
> 	$cnt += 1
> 	...
> 	--
> 	-- you should do FULL backups once a week or once a month ???
> 	--
> 	-- there are major problems with doing just 24hr incrementals
> 	-- you should perform incrementals since last FULL backup...
> 	--
> 
> -- if you are worrying about intruders etc... you now have 3 boxes
>    to worry about instead of just one
> 
> -- you should encrypt your backups if security is an issue
> 
> -- once a month...erase your old backups from few months ago... via cron
> 
> -- one day, when time is available...create a new test file...
> 	- backup it up by manually doing the backup....
> 	- pretend you just have a major primary disk failure
> 	- and try to restore everything from backup
> 
> -- now you can rest assured you have a reasonable backup ??
> 
> On Fri, 18 May 2001, Miquel Mart?n L?pez wrote:
> 
> > Hi all,
> > I am trying to decide on backup schecduling and have several things in
> > mind. My system is on a main hard disk, and backups are made by copying this
> > HD to another (equal sized) one. 
> > 
> > Now, the point is I have one main HD and 2 backup HD, and have two criteria:
> > 
> > a) Backup every one or two days, alternating the backup disks, so that the
> > oldest backup is max 4 days old, and the newest is just 2 
> > 
> > b) Backup every one or two days with the same backup disk, and make weekly
> > backups with the other one.
> > 
> > As you see, it's a compromise between having updated backups and having the
> > time to realize problems or intrusions.
> > 
> > So, just wondering, what do you do/suggest?
> > Thanks!
> > Miquel Martín
> > 
> > 
> > -- 
> > To UNSUBSCRIBE, email to debian-user-request@lists.debian.org 
> > with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
> > 
> 



Reply to: