Re: best practice for hard drive upgrade

ok, I know this thread is now getting a little old, but what was suggested by
SuperGumby did not work. I pulled 1 250, added a 500, rebuilt array then
pulled 2nd 250, rebuilt array, and ended up with a 250gb array.
Apparently this RAID card - promise tx4310 - will not resize the array.
Windows disk management did not see the blank space on the drive, so there
was no way to take control of that space. I flirted with the idea of using a
partition manager, but decided that since its not really supported, I won't
do it.

So its off to restore from backup to a fresh array.

I did read somewhere that this techinque does work on a FastTrak S150 SX4.


"SuperGumby [SBS MVP]" wrote:

hmmm, interesting that the Promise tech suggested that. The TX4310 data
sheet suggests:
Online Capacity Expansion - add capacity on the fly
RAID Level Migration - change array levels on the fly
http://www.promise.com/marketing/datasheet/file/FT_TX4310_DS_032206_web.pdf

but we hit a couple of ideas about how RAID controllers handle such change.
1) larger drives get treated as same size. I haven't seen this since using
an older HP SCSI RAID. If the original array consisted of 36GB drives a,
say, 73GB drive could be used but the controller actually formatted the
drive as 36GB. This refers NOT to Windows level 'format' but to LLF (Low
Level Format), to all intents the larger capacity drive was converted to
lower capacity.
2) apparent size of array and RAID volumes. RAID volumes are not to be
confused with Windows volumes. To a degree modern controllers hide most of
this from you. The controller also may not support multiple volumes.

Enter the wayback machine and we have a RAID controller and a bunch of
drives (spindles). Connect it, fire it up, you have no available drive
space. You must first create your RAID set(s), tell the controller that it
should use the spindles. This is so far back that I'm actually getting
sketchy on the details, I'll go with what I think is right. At this point
you still have no space available for use, you must create RAID volumes, at
which time space is still not available, until you create RAID drives. The
array is now at a point where (each) RAID drive becomes available to the OS
(or install). We might have 10 spindles in the set, two RAID volumes, and
four RAID drives.

(for pity's sake Mick, get to the point)

I haven't used the TX4310 but doubt it performs as described in 1 above. For
the most part because current SATA HDD's will not (TTBOMK) allow such
operation. I _expect_ that if you built a RAID1 from your existing 250GB
drive to a 500 that it would more simply just use 250GB of space and leave
the rest unused. You could then remove the 250 and replace it with a 500,
rebuild the array. At this point we have two 500's in RAID1 supplying a
single RAID set/volume/drive of 250GB capacity and having unused space. You
may be able to a) create another RAID volume in the unused space or b) use
online capacity expansion to consume the space.

In either a) or b) the existing Windows partitions will not be adjusted, and
I suspect this is what the Promise tech may have suggested, however the
extra space will either be available for use (to the OS) as an apparently
new drive OR unpartitioned space on the existing drive.

It's worth a go. Step 1 is full backup and step 2 is to remove one of the
existing 250's, giving you a fallback point if all this is some madness
running around my head. Stop SMTP coming in and shutdown all attached
systems.
Shut down and remove drive0 (existing 250). Put it in a static bag out of
harm's way, this is our fallback.
Plug in one of your 500's and rebuild the array between the 250 and it.
When rebuild is complete inspect the aray management, what does it say about
the attached drives? Does it offer unused space? We don't actually want to
do anything at this point but knowing how the controller is handling the
drives _may_ be of benefit.
Shutdown and replace the 250 with your 2nd 500. Rebuild the array and
inspect array management, what does it now say about the drives?
Go into both the RAID Level Migration (not likely to say anything) and
Online Capacity Expansion areas of the management interface, is the
additional space available for use in any manner? (either as additional
space for the existing array or as free space for a new array).

If we can't 'migrate' in this manner simply shut down and disconnect all
drives, return Drive0 (the one on the shelf) to service, start the system,
shutdown, re-eastablish your (250) array, and we look for PLAN B.

The advantage here is that we do everything from the RAID controller. The
additional space we create becomes an additional drive or partition to the
OS, no 'imaging' or 'partition adjustment' is performed so the system
remains in a 100% MS supported state.

Guaranteed? no.
Worth a try? Up to you.

--
SBS remote support services. (Fees apply)
mickm at mickmalloy dot dyndns dot org

"M. Murphy" <MMurphy@xxxxxxxxxxxxxxxxxxxxxxxxx> wrote in message
news:1D98DCBD-5386-4733-98CB-CEC702FD4224@xxxxxxxxxxxxxxxx
Thanks Larry,
That sounds like a very logical plan, my backups (according to the log
file)
take a few hours including the verify part. I guess this means the server
and
email will be offline for most of the day.

I spoke to Promise support just now, and the tech there assured me that I
can not reclaim that extra space if I put the bigger drives in and rebuild
the array, so its looking like your method is my only real choice. I will
have the old drives to fall back to if there is a problem.

Thanks for your support!

"Larry Struckmeyer [SBS-MVP]" wrote:

While that would work, I suggest this is a perfect change to test your
backup/recovery plans. Make a good backup. Test it. Remove the
existing
drives, install the new ones, create the array, partition it however you
like, so long as each partition is as large or larger than the old one.
Restore from backup according to this document:

http://www.microsoft.com/smallbusiness/support/articles/backup_restore_sbs2003.mspx


--
Larry
Please post the resolution to your
issue so that others may benefit.


"M. Murphy" <MMurphy@xxxxxxxxxxxxxxxxxxxxxxxxx> wrote in message
news:9734A58B-30CD-4470-B3B6-5D4218A6BE87@xxxxxxxxxxxxxxxx
Thanks for the reply Larry,

here is my setup - SBS 2003 R2
I have 2 250gb drives in a raid 1 setup.
Raid is hardware, and its a promise tx4310 card
The "logical drive" is partitioned - c: is for the system and its 30gb,
f:
is about 203gb and exchange and all the other data is on the larger non
system drive.

May I run this scenario by you and everyone else:
1) I pull out 1 of the old 250gb and replace with 500gb, then rebuild
the
array

2) then pull out the other 250 and replace with 500 and rebuild again

could I then make another logical drive out of the unused space (with
the
disk management console)? If that would work, I think it would be less
painful and less stressful, and I would still have the old drive to put
back
in, in case of something going haywire. This would also not require
any
3rd
party software.

I have been reading about using acronis true image to do this , and its
scaring me into not using it. The price tag is also pretty
frightening!!



"Larry Struckmeyer [SBS-MVP]" wrote:

Hi:

What version of SBS? Windows 2008 allows for resizing the partitions.
Also, you could use 250 + 250 and not mess with the backup restore.

However, to answer your question, the only fully MS supported way is
backup
the old drives, install the new ones, create the array, install CD1,
install
the appropriate Windows Service Pack, and restore from backup.

Having said that, Storage Craft and Accronis have booth been mentioned
here
with success. If you are brave, there are third party utilities that
would
allow you to extend the 250 to 500 once the new drives are in place.

I hope these are data drives and not system drives, as you should be
separating the system and programs (c:\) from user data (d:\ or
whatever
is
next available).

--
Larry
Please post the resolution to your
issue so that others may benefit.


"M. Murphy" <MMurphy@xxxxxxxxxxxxxxxxxxxxxxxxx> wrote in message
news:FA071495-58B6-47F6-BD00-BAC47F1361CA@xxxxxxxxxxxxxxxx
Hi,
Anyone here have any info on upgrading hard drive size?
I am using a raid 1 setup, my WD 250gb drives are approaching the
end
of
their expected lifetimes. Boss bought WD 500gb drives for server.
They
are
all enterprise WD5002abys drives.
We had thought that we could just put the bigger drives in 1 at a
time
and
rebuild the array, but found out that the end result would still be
a
250gb
drive.
Looking into acronis to image the original drives, anyone have any
gotchas
to look out for when doing this? Is this the best way to go?

Thanks to all!!






.

Relevant Pages

  • Re: Problems with software RAID on SATA
    ... Connected to this are two 320GB drives ... >>which I want to turn into a RAID1 array. ... >>I'm almost certain it's a problem with initting the RAID arrays at boot. ...
    (Debian-User)
  • Re: RAID newbie...can I have several partitions on a RAID 1 array?
    ... You haven't expounded upon why you think you need raid. ... better backup device rather than buy 2 cheap RAID HBAs. ... RAID array then I would have to replace the mobo with the same one or at ... Lets say, for example, you buy 2 identical model drives, from ...
    (comp.sys.ibm.pc.hardware.storage)
  • Re: [PATCH 000 of 5] md: Introduction
    ... "why linux raid isn't Raid really, why it can be worse than plain disk") ... After this, the array ... error is in the filesystem, due to the complex layout of raid5. ... hundreds or 1000s of drives, you've quite high probability that some of them will fail sometimes, or will develop a bad sector etc). ...
    (Linux-Kernel)
  • Re: LSI Logic SATA Raid Restore
    ... The drives are identical. ... AFAIK it is a LSI Logic Raid chip - It's a Fujitsu Siemens Econel 40 Server. ... I did a "rebuild" from the BIOS menu. ... I am assuming that the array is not properly rebuilding and the data is not ...
    (comp.sys.ibm.pc.hardware.storage)
  • Re: Paul and Old Man: Cannot fix RAID5 failure ...
    ... all of the sudden the Intel Matrix Storage ROM showed Rebuild status. ... When booting form RAID, there is some activity while the screen is black, ... Considering 3 SATA drives spining up after powerdown may cause a 12V ... I was considering a parallel WinXP installation on a 4th disk, ...
    (alt.comp.periphs.mainboard.asus)