Re: Intel Matrix Storage... how do I rebuild a degraded RAID10 array?

In article <m31wszwugp.fsf@xxxxxxxxxxxxxxxxxxxxxxx>, Ronald Cole
<ronald@xxxxxxxxxxxxxx> wrote:

I put four Maxtor MaxLine III SATA drives on the ICH7R interface of my
ASUS P5WD2-E Premium mainboard. I configured them in a RAID10 array
before installing Windows XP. Well, I suffered infant mortality on
one of the drives and Intel Matrix Storage Manager dutifully informed
me that the drive went missing and that the array was running in
degraded mode.

So I install a new drive where the dead drive was and the <ctrl-I>
BIOS says that the new drive is not a member of the raid array, but
doesn't have any obvious way to let me tell it that it's the old
drive's replacement.

So I boot into Windows and the Intel Matrix Storage Console shows that
the one drive of the RAID10 array is still missing and that there is a
new non-RAID drive, but apparently no obvious way to tell it that it's
the old, dead drive's replacement and to immediately commence with
rebuilding.

I've emailed Intel's Matrix Storage support, but they still havn't
gotten back to me.

How do I fix this? What ASUS or Intel document actually explains how
to replace a dead member of a RAID array?

Thanks in advance

In previous reading on the subject, it seems the user is supposed
to take a new disk and mark it as a "spare" or hot standby. Changing
the status of a new disk from <unknown> to "spare", tells the software
that the user wants the "spare" to be used for RAID purposes.
That is how you would do it on non-Intel RAID controllers.

I would fire up the Matrix Storage Manager, find whatever
fancy graphical view it has got, highlight the new disk
and see if it has properties. Maybe there is some popup menu
that allows the properties to be specified or changed ?
The manual certainly doesn't tell you.

It is funny, but the Intel manual for this, basically makes
it seem like osmosis - the MSM just grabs the new disk
and knows immediately that the user wants to rebuild their
busted array with it. Which is just crap. The glossary of
the Intel manual, even takes pains to explain what a "spare"
is, but the word spare is not used in the rest of the manual.

You are welcome to marvel at their latest effort here (pg.71):
ftp://download.intel.com/support/chipsets/imsm/sb/manual57_oem.pdf

Paul
.

Relevant Pages

  • Re: Hard drive question
    ... 146Gb array. ... >> More drives is theoretically faster, but I doubt you'll notice it in ... >> Please consider a hot spare drive, even if you add it later. ... >> the server, and it'll be covered on the server's warranty. ...
    (microsoft.public.windows.server.sbs)
  • Re: Borked MD RAID...
    ... /dev/sdd1", drives no longer are allocated as spares, however the array ... currently the array is being rebuilt. ... spare. ... Next verify that the UUID numbers in your /etc/mdad.conf file stored in the initrd file is correct, you'll have to extract the initrd file ...
    (Fedora)
  • [Summary] saving disksuite info while rebuilding server
    ... "Caution - If a full install of Solaris is performed, ... make sure you have tape backups of everything. ... I had an old workgroup storage array with six 18G drives formatted RAID5 ...
    (SunManagers)
  • Re: Testing Spare Drives in Software RAID
    ... Generate a TestMessage alert for every array found at ... This alert gets mailed and passed to the alert program. ... to test the spare drive in an array. ... array has two failed drives and the array itself fails. ...
    (Debian-User)
  • Re: best practice for hard drive upgrade
    ... pulled 2nd 250, rebuilt array, and ended up with a 250gb array. ... Apparently this RAID card - promise tx4310 - will not resize the array. ... If the original array consisted of 36GB drives a, ... Plug in one of your 500's and rebuild the array between the 250 and it. ...
    (microsoft.public.windows.server.sbs)