Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.debian.user > #188819 > unrolled thread

How to Clone a RAID1 Box to Another with Drives Twice As Big?

Back to article view | Back to linux.debian.user

Contents

  How to Clone a RAID1 Box to Another with Drives Twice As Big? Kent West <westk@acu.edu> - 2017-11-09 19:30 +0100
    Re: How to Clone a RAID1 Box to Another with Drives Twice As Big? Roberto C. Sánchez <roberto@debian.org> - 2017-11-09 19:40 +0100
      Re: How to Clone a RAID1 Box to Another with Drives Twice As Big? David Christensen <dpchrist@holgerdanske.com> - 2017-11-09 20:10 +0100
      Re: How to Clone a RAID1 Box to Another with Drives Twice As Big? Michael Stone <mstone@debian.org> - 2017-11-09 20:40 +0100

#188819 — How to Clone a RAID1 Box to Another with Drives Twice As Big?

[Multipart message — attachments visible in raw view] — view raw

I have a Debian Stable box with two 1TB drives, set up as a mirrored RAID1
setup.

I have a replacement box with two 2TB drives.

How do I best clone the old machine to the new, with minimal (or better,
no) downtime, being able to wind up with double the drive space on the new
box?

Thanks!

-- 
Kent West                    <")))><
Westing Peacefully - http://kentwest.blogspot.com

[toc] | [next] | [standalone]

#188820

On Thu, Nov 09, 2017 at 12:25:17PM -0600, Kent West wrote:
>    I have a Debian Stable box with two 1TB drives, set up as a mirrored RAID1
>    setup.
> 
>    I have a replacement box with two 2TB drives.
> 
>    How do I best clone the old machine to the new, with minimal (or better,
>    no) downtime, being able to wind up with double the drive space on the new
>    box?

If minimizing downtime is your objective, then I would recommending
doing a fresh install on the replacement system and rsyncing everything
over.  That lets you have the old system up and running until the very
last moment.

Of course, if you want a true clone (i.e., block for block disk image),
then I would recommdend shutting down the old machine, plugging the old
drives into the new machine, booting a live image, and then transferring
everything from each old drive to its new counterpart with dd.

Once the dd operations are complete, then you can either create new
partitions to use the free space and then assemble the new pair of
partitions into a new /dev/md? device that you can use.  If you are also
using LVM, then you add the new /dev/md? as a physical volume to your
system's volume group.  That way you would be able to create new logical
volumes.  Alternatively, you could resize the last partition on each
disk to take advantage of the additional space.  However, the
suitability of that approach depends on your current partition scheme.

Regards,

-Roberto

-- 
Roberto C. Sánchez

[toc] | [prev] | [next] | [standalone]

#188823

On 11/09/17 10:33, Roberto C. Sánchez wrote:
> On Thu, Nov 09, 2017 at 12:25:17PM -0600, Kent West wrote:
>>     I have a Debian Stable box with two 1TB drives, set up as a mirrored RAID1
>>     setup.
>>
>>     I have a replacement box with two 2TB drives.
>>
>>     How do I best clone the old machine to the new, with minimal (or better,
>>     no) downtime, being able to wind up with double the drive space on the new
>>     box?
> 
> If minimizing downtime is your objective, then I would recommending
> doing a fresh install on the replacement system and rsyncing everything
> over.  That lets you have the old system up and running until the very
> last moment.

+1 for a file server.


If it's a workstation, put all the drives in the fastest box, mount both 
RAID's, and then rsync.


David

[toc] | [prev] | [next] | [standalone]

#188826

On Thu, Nov 09, 2017 at 01:33:51PM -0500, Roberto C. Sánchez wrote:
>On Thu, Nov 09, 2017 at 12:25:17PM -0600, Kent West wrote:
>>    I have a Debian Stable box with two 1TB drives, set up as a mirrored RAID1
>>    setup.
>>
>>    I have a replacement box with two 2TB drives.
>>
>>    How do I best clone the old machine to the new, with minimal (or better,
>>    no) downtime, being able to wind up with double the drive space on the new
>>    box?
>
>If minimizing downtime is your objective, then I would recommending
>doing a fresh install on the replacement system and rsyncing everything
>over.  That lets you have the old system up and running until the very
>last moment.

It also gives you a straightforward fallback plan. The other way to do 
this is to add the two new 2TB disks to the mirror, remove the 1TB 
disks, then grow the md to the full size of the 2TB disks. The upside is 
that it can all theoretically be done without even a reboot if you have 
hot-swap drives. The downside is that if it breaks, you're left holding 
both pieces and have to start over from backup. (There are some fiddly 
bits I glossed over as well, like making sure the partitioning is 
correct, booting works, etc.) So "best" here depends on your 
constraints.

Mike Stone

[toc] | [prev] | [standalone]

Back to top | Article view | linux.debian.user

csiph-web