Do you have a question? Post it now! No Registration Necessary.  Now with pictures!

Threaded View

I am guessing I am screwed.

I have a Windows Server that had RAID 0 on it (I
didn't set it up that
way, previous tech guy did). It is a ATA/133 RAID built
into the

I got a message from the server saying that there is not
enough disk
space. So, I slid off the side panel on the case and peered inside.
There was a CDROM with an extra spot for an IDE hard drive off it. So
I decied
to throw a Western Digitial drive on it (Secondary
controller). I shut the
computer down first of course, quickly
attached the second drive, and rebooted.
First message that I got on
the reboot (after the BIOS) was "Invalid Partition."
Uh-oh. I thought
that the WD I just put in was thinking it was the boot device.
checked the BIOS, and no, it was not--HDD0 was, and that is from the
Rebooted, now just a blank black screen.

Scared, I completely took the WD out.
Rebooted, same thing. Checked
the RAID drives, still connected, still spinning
up. Rebooted, and
now I get some kind of "trap" code every time I boot with a
bunch of
meaningless (to me) numbers.

Unfortanately, there is no backup because
the company does not have
the funding for one.

Is there any way possible to
completely rebuild this RAID 0 or get the
data off it in some way? Pretty
please? Help? Suggestions? I am going
to try and boot with a Windows CD and the
RAID drivers and see if it
sees anything on the RAID. Hopefully it will, but the
trap code
worries me.


On Thu, 08 Sep 2005 19:32:04 GMT, (winux) wrote:

Quoted text here. Click to load it

This seems to be your mistake, that the bios was set to boot
from the onboard IDE first.  You need to go back into the
bios and recheck all boot-related settings.

Quoted text here. Click to load it

You could try to access the raid controller's setup screens
and survey which drives are enumerated as members of the
raid, but it shouldn't be necessary.  Do not change any
settings in the RAID setup.

Quoted text here. Click to load it

It is a necessary business expense.  There is no such thing
as no funding.  You need to backup the data, confirm it's
restorable from the backup (a test-run of the backup
strategy confirming it works) even a 2nd copy of the backup
if this is critical data, then get rid of the RAID0 arry.
Make it RAID1 or 5 or whatever, not RAID0.

Quoted text here. Click to load it

No, you can't rebuild a bad RAID0.  It shouldn't be bad
though, your actions (as described) would not have
jeopardized the RAID0 array... it is, as described, only an
issue of setting the motherboard bios correctly to boot from
the RAID controller first.  This presumes there are no
cables or cards jostled loose during the installation of
that hard drive or afterwards... reexamine the system to see
if anything else has been disturbed.



After tinkering it for a while, I was finding I could not boot from
anything--not even a floppy or CD as it would just hang; even with
the RAID
controller disabled, and basically just the motherboard,
video card, and floppy
disk attached. Even setting the BIOS to
failsafe settings (no default option)
didn't work. If I were to leave
it off for awhile, the system would begin to
boot, but then either
freeze or go to a blue screen of death (random errors).

So, I decided to flash the BIOS, but before I did I read the
instructions on the
CMOS clear jumper. So, I did that first before
the flash, and then zam! it
worked. So, something must have been in
the BIOS to keep the system from booting
at all.

Upon booting, I immediately backed up all data. The server does not
have a virus scan (they don't want to pay the money for one), but I
am thinking
that perhaps the BIOS was hit and my just rebooting it
after weeks of being
constantly on kicked everything into motion.

Now that this scare happened, the
boss finally realizes the importance
of a backup system and virus scan. However,
we can skip the virus scan
since we are moving to a Linux server instead...with
a mirroring RAID.

Site Timeline