9fans - fans of the OS Plan 9 from Bell Labs
 help / color / mirror / Atom feed
From: steve jenkin <stevej098@gmail.com>
To: Fans of the OS Plan 9 from Bell Labs <9fans@9fans.net>
Subject: Re: [9fans] OT: how do 9fans backup their Mac(book)?
Date: Sun,  2 Oct 2011 13:28:42 +1100	[thread overview]
Message-ID: <4E87CC5A.1090006@gmail.com> (raw)
In-Reply-To: <906e80d587e99006633899b8d56be210@brasstown.quanstro.net>

w.r.t. Disk Cloning.

you are conflating multiple issues:
 - hardware disk imaging
 - managing NAS appliances
 - handling VM images

I agree with you.
Cloning drives in low-end NAS appliances is quick, simple and effective.
I had a lot of fun a year ago documenting how to clone config and data
from one Thecus to another.

But that approach doesn't/won't work on high-end systems or 'enterprise
grade' LVM systems, like Veritas, and the IBM/HP LVM system.
They build in too many checks, because they give you all the tools you
need to move data around and keep it 'known' by their system.

How many times have you cloned a physical disk yourself, NOT in a NAS,
but on Apple hardware for OS/X?
If you haven't done exactly that, you have no idea of the magnitude of
the problem, or if it is feasible at all.

After having to clone disks in systems far too many times, I've come to
find it tiresome and time wasting.
If the situation demands you do it, you've already failed as an Admin...

At home, I have an 80MB (m-e-g-a b-y-t-e) drive in an old PC that I want
to access.
Really, I'd like to access the data, so how do I do that?
And for some extra fun, upgrade the disk on that old existing system so
I have more space available.

cloning the 80Mb onto a more recent drive won't work.
The BIOS notices the mismatch and won't play.
More significantly, if I could bodgy-up everything so the new drive
worked, I'd still only have 80Mb and no way to expand it.
The disk partitioner respects the limits of the drive. If I simply 'dd'
the drive onto another, it fails.

Couldn't take an image of the drive either: IDE version was so old,
modern adaptors wouldn't work.
So I figured out how to transfer the files to a 2nd, newer, drive with a
large filesystem whilst mounted in the PC.
[The system would only allow a maximum filesystem of 472Mb. The rest of
the drive was wasted]

Having a readable copy of the data, NOTE, not the original
hardware-disk-image, I've taken a copy of that filesystem so I can run
up a VM in the future. That VM image I can now 'clone' to my hearts content.

You are right that cloning disk *images* is an essential part of
using/managing VM's.
Don't conflate that with cloning physical disks.

Physical cloning is not something to trifle with when you're looking
after Other People's Data/Systems, more a last-chance desperation
measure. [see above. having to clone means everything else has failed]

Something we're going to see more of is the physical disk information
being incorporated into a motherboard-based security system.
We are going to see more "Trusted Platform Module" systems shipped - and
they notice little things like changing the drive-id...

As well, Apple adopted GPT drives. Not the 'industry standard' disk
partitioning.
I simply don't know enough to fiddle everything that's needed to support
both arrangements - there's a lot of fine detail to get right...

It's not unusual to want or need to mount the original and copy drive in
the same machine.
If you've cloned the GPT, that's not going to happen.

Have you ever tried to move a Windows Vista partition? There are quite a
few subtleties & 'gotchas'.
I spent around a month, near full-time, on this problem several years
ago and bounced off it.
I've a whole bunch of experience with FAT, NTFS, Unix, Linux, multiple
LVM systems filesystems, disk-arrays or not, and their supporting disk
partitioning, but I couldn't get this to work...
I've dug people out of seemingly impossible situations many times, but I
didn't crack that one - nor had anyone else from what I was reading on
the web at the time.

It's trivial to clone/migrate & resize an NTFS partition with Win-XP and
below, but Vista had a few too many checks.
Doesn't play well with 'grub', either.

erik quanstrom wrote on 2/10/11 12:28 PM:
> really?
>
> what do you mean by "doesn't allow for hardware changes"?
> i
> also, i'm pretty sure that whenever the new disk is larger
> than the end of the last used partition, both fdisk and prep
> just see more empty space at the end of the partition.  judging
> from the spec, gpt should work the same way.  where do you
> see cloning onto a larger disk failing?
>
> cloning a disk is really common for nas appliances, and it's
> not too hard to make your storage whatever size you require.
>
> most advanced storage features of vmware work by cloning
> storage.  i see the the idea of cloning disks becoming more
> prevalent.
>
> - erik
>


--
Steve Jenkin, Info Tech, Systems and Design Specialist.
0412 786 915 (+61 412 786 915)
PO Box 48, Kippax ACT 2615, AUSTRALIA

stevej098@gmail.com http://members.tip.net.au/~sjenkin




  reply	other threads:[~2011-10-02  2:28 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-09-30 15:19 Axel Belinfante
2011-09-30 15:53 ` Anthony Sorace
2011-09-30 17:43   ` Lyndon Nerenberg (VE6BBM/VE7TFX)
2011-09-30 17:59     ` erik quanstrom
2011-09-30 19:21       ` Lyndon Nerenberg (VE6BBM/VE7TFX)
2011-09-30 19:42         ` erik quanstrom
2011-09-30 19:45           ` Lyndon Nerenberg (VE6BBM/VE7TFX)
2011-09-30 19:53             ` erik quanstrom
2011-10-01  8:21       ` Anthony Sorace
2011-10-01 15:14       ` Ethan Grammatikidis
2011-10-01 15:24         ` erik quanstrom
2011-10-01 15:48           ` Ethan Grammatikidis
2011-10-01 16:49             ` Venkatesh Srinivas
2011-10-01 18:21         ` Lyndon Nerenberg (VE6BBM/VE7TFX)
2011-10-01 18:29           ` Anthony Sorace
2011-10-02 15:26             ` Ethan Grammatikidis
2011-11-03 20:54       ` Michaelian Ennis
     [not found]       ` <CAAQydVjZ42C-GTEFT7E2t=dCQ35fvHzqjSYx9a3o+8v+5D1vfA@mail.gmail.c>
2011-11-03 20:57         ` erik quanstrom
2011-11-01 18:15   ` smiley
2011-11-01 18:57     ` Anthony Sorace
2011-09-30 19:50 ` Latchesar Ionkov
2011-10-01 15:21   ` Ethan Grammatikidis
2011-09-30 20:01 ` Francisco J Ballesteros
2011-09-30 21:15 ` Bakul Shah
2011-10-01 17:22 ` John Stalker
2011-10-02  0:35 ` steve jenkin
2011-10-02  1:28   ` erik quanstrom
2011-10-02  2:28     ` steve jenkin [this message]
2011-10-02  2:37       ` erik quanstrom
2011-10-02  3:07         ` steve jenkin
2011-10-02 12:38           ` Iruatã Souza
2011-10-03  3:57             ` steve jenkin
2011-10-03  4:02               ` erik quanstrom
2011-10-02 13:30           ` Salman Aljammaz
2011-10-03  3:47             ` steve jenkin

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4E87CC5A.1090006@gmail.com \
    --to=stevej098@gmail.com \
    --cc=9fans@9fans.net \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).