Re: Full restore - with time to plan it

Thank you all very much for your input and sorry not to reply sooner. I am
assuming now that I will have backed up the bad data in the bad stripe. My
current SBS installation is working quite well at the moment and I'm hoping
that any corruption is in a far away corner of the user data, rather than the
working server. I don't see any alternative to biting the bullet and doing
what I proposed:
backup>rebuild>reinstall OS>restore
Which still makes me nervous; still I take it there are no gotchas.
Thanks again.

"Anna Clark" wrote:

Hey Steve:

I wondered if anyone would read that to its end.

Thanks for the comment, I think? <BG>

Anna

"Steve" <newsgroup@xxxxxxxxxx> wrote in message
news:uLy4V6lPHHA.5012@xxxxxxxxxxxxxxxxxxxxxxx
Anna-well said (verbosely) -:)

"Anna Clark" <anna.clark(remove this)@verizon.net> wrote in message
news:uxVvIzVPHHA.2312@xxxxxxxxxxxxxxxxxxxxxxx
Hi Super Hi Sorehead:

Super has cleared up my confusion re the problem with the array, but
neither
of us have answered the question re precautions to take before wiping
the
drives, breaking the array and recreating. I tried to condition my
first
reply in such a way that 1. if it did in fact require breaking the
array,
not replacing a drive, and 2. if an image was unlikely to be valid,
and
3.
if I/we knew more about the available resources, we could offer
suggestions
that might be useful to anyone in this position.

Since I don't yet know anything about the available resources, I still
don't
have a better answer than the one Sorehead proposes. But from a casual
read
of the info given by Super, it appears that an image is unlikey to be
valid.
This raises the question of what backups are likely to be valid, and if
the
OP knows when the bad stripe table reached its max, and what caused the
table to fill in the first place.

BTW, I did Google for the array stripe failure condition, but had so
many
hits that I decided to ask about it from the person experiencing it,
rather
than filter and re filter the results. Good searching, Super.

And finally, I have often written very long involved answers (admittedly
sometimes convoluted) to questions that seemingly failed to engage the
op,
and have taken a lessen from Cris, even if his method is not
intentional.
He very often asks a short question in response to a post, and if he has
the
op's attention, then gives the long(er) - Cris is not as verbose as
some-
answer.

--
Regards:

Anna Clark
-----
Please do post the conclusion or solution
to your issue so that others may benefit.


"SuperGumby [SBS MVP]" <not@xxxxxxxxxxx> wrote in message
news:%23FFiQKUPHHA.4280@xxxxxxxxxxxxxxxxxxxxxxx
something Freudian there Anna ;-)

A stripe error is a (hopefully) rare condition where the array data...
maybe
I'll just google...
http://www-1.ibm.com/support/docview.wss?uid=psg1MIGR-57140

Replacing drives in the array does not necessarily clear the condition
(actually, it's unlikely to). In most cases the array needs to be
cleared
and recreated.

"Anna Clark" <anna.clark(remove this)@verizon.net> wrote in message
news:eQn01qMPHHA.448@xxxxxxxxxxxxxxxxxxxxxxx
Um, make that a "stripe" error. Don't know for sure what that is
either.
:-)

Anna

"Anna Clark" <anna.clark(remove this)@verizon.net> wrote in message
news:upCL6gMPHHA.5012@xxxxxxxxxxxxxxxxxxxxxxx
Hi:

Don't know for sure what a strip error is, (failure of one drive?)
If
it
is a failure of one drive, you should be able to replace just that
drive.
If this is not what it is, it sounds serious enough that an image
may
not
be reliable? How much data on the array, and is it the only
drive/drive
set in the box?

Any external drives, or additional drives you an put in the box
available. Don't have to be on the same controller, but could be.

--
Regards:

Anna Clark
Please do post the conclusion or solution
to your issue so that others may benefit.


"Sorehead_5" <Sorehead5@xxxxxxxxxxxxxxxxxxxxxxxxx> wrote in message
news:F6FDC30C-EDC8-4B87-8790-8F4403781DE5@xxxxxxxxxxxxxxxx
I have an IBM 3 element RAID 5 array. Recently it has developed a
stripe
error. IBM say that I should rebuild the array. This involves
backup>rebuild>reinstall OS>restore.
It makes me nervous. In theory, of course, I can do this already
but
still....
Anyone ouut there with any reassurance and tips? Should I make an
ASR
backup
or a Normal Full backup and reinstall Server 2003.? Do I need to do
anything
more?
I'm running SBS 2003 SP1.
Any input gratefully received.
Thanks.













.

Relevant Pages

  • Re: SBS2003 rebuild
    ... I agree he should just image the server using an appropriate WinPE ... disk, wipe the array, create and initialize a new array (and allow the ... ServeRAID to completely scrub the drives), ... A bad stripe isn't the result of a single drive failure....something ...
    (microsoft.public.windows.server.sbs)
  • Re: [opensuse] Re: [OT] vmware and fake scsi devs
    ... Single threaded access to a raid array may not be helped by adding ... by adding disks to an array. ... So if you have M drives, ... stripe, then the existing parity data can be ignored. ...
    (SuSE)
  • Re: IOPS from RAID units
    ... IOPS the array can deliver is the aggregate of all the drives in the raid ... I write 5 bytes to a raid5 array of 10 drives. ... probably more depending on the stripe size of the raid array. ...
    (comp.arch.storage)
  • Re: Full restore - with time to plan it
    ... Super has cleared up my confusion re the problem with the array, ... BTW, I did Google for the array stripe failure condition, but had so many ... Replacing drives in the array does not necessarily clear the condition ...
    (microsoft.public.windows.server.sbs)
  • Re: SBS2003 rebuild
    ... I agree he should just image the server using an appropriate WinPE ... disk, wipe the array, create and initialize a new array (and allow the ... ServeRAID to completely scrub the drives), ... A bad stripe isn't the result of a single drive failure....something ...
    (microsoft.public.windows.server.sbs)
Loading