Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 6 Mar 2000 02:11:47 -0800
From:      "Paul M . Lambert" <plambert@plambert.net>
To:        Matt Heckaman <matt@ARPA.MAIL.NET>
Cc:        FreeBSD QUESTIONS <questions@FreeBSD.ORG>
Subject:   Re: dump/restore question..
Message-ID:  <20000306021147.I766@pinky.plambert.net>
In-Reply-To: <Pine.BSF.4.21.0003060232090.10633-100000@epsilon.lucida.qc.ca>
References:  <Pine.BSF.4.21.0003052316090.11594-100000@boris.netgate.net> <Pine.BSF.4.21.0003060232090.10633-100000@epsilon.lucida.qc.ca>

next in thread | previous in thread | raw e-mail | index | archive | help
Just a friendly reminder, as I've done this exact thing recently, and
ran into a few problems I could have avoided.

1) be sure that your kernel is ready for the extra drive before you
reboot, so you only have to reboot once. i.e. make sure you have the
right number of drives and controllers figured in, if you're using IDE.

2) shutdown the machine, add the drive, and boot to single user.  make
sure the drive was probed correctly.  if so, mount -o update,rw /
and mount -a to get all the partitions mounted and exit to start
into multiuser.

3) once in multiuser, make the new partitions and copy your data
over.  you may have to do this in single user if your data is being
modified while you copy it, or else arrange for nothing to be
updated while the copy occurs.  worst case, you might want to copy
the data over, then go back to single user and do an rsync across
the drives.

4) make sure your /etc/fstab is correct for the _new_ drive being
the only one.  also make sure that the new drive has boot blocks.

5) bring the box down again, remove the old drive, move the new drive
if necessary, and cross your fingers.

6) bring the box back up.

7) fix your mistakes.  ;-)

on a production box, i might sometimes bring it up with both drives
still attached, but booted from the new one.  that way, if there's a
problem, i can mount the old drive and get to the data there.

then after a while when it's clear it's ok, i'll bring it down again
to get the old data.

this is assuming you want multiple short outages instead of one long
one; in some production situations, the latter is preferrable.

--plambert



On Mon, 06 Mar 2000, Matt Heckaman wrote:

> That is a wonderful idea, I never thought of doing that, sure saves a fair
> amount of time, in theory that could be done with minimum downtime. 
> 
> Matt
> --
> Matt Heckaman [matt@arpa.mail.net|matt@relic.net] [Please do not send me]
> !Powered by FreeBSD/x86! [http://www.freebsd.org] [any SPAM (UCE) e-mail]
> 
> On Mon, 6 Mar 2000, wellsian wrote:
> 
> : Date: Mon, 6 Mar 2000 02:28:17 -0500
> : From: wellsian <wellsian@caffeine.com>
> : To: Matt Heckaman <matt@ARPA.MAIL.NET>
> : Cc: FreeBSD QUESTIONS <questions@FreeBSD.ORG>
> : Subject: Re: dump/restore question..
> : 
> : You'll be fine. The restore program writes files, not disk blocks like dd.
> : It behaves in the same manner as ftp or other user-programs, and doesn't
> : care about the size of the original filesystem. (assuming there's enough
> : space of course)
> : 
> : If you have the opportunity, I would attach the replacement drive to the
> : same system and perform the dump/restore process in one shot. Or better,
> : use cpio or pax. That would save lots of time over a tape or network dump
> : and eventual restore, though it sounds like a full backup before starting
> : might be wise. :)
> : 
> : Dave
> 
> 
> 
> To Unsubscribe: send mail to majordomo@FreeBSD.org
> with "unsubscribe freebsd-questions" in the body of the message
-- 
I hate bombs, terrorism, fear, plans, future and past injustices, manifestos,
popular sentiment, ignition, timetables, meetings, and poorly adjusted
weasels.  A warm hello to my friends and fans in domestic surveillance!


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000306021147.I766>