Re: URGENT: Need help rebuilding iir RAID5 array with failed drive


On Thu, 2008-06-26 at 06:49 -0700, Garrett Cooper wrote:
Hello,
First off sorry for the cross-post. I typically don't do this
but this is an important question, so please bear with me. I'm just
trying to get more eyes on the subject so I can (maybe) get a reply
quicker...
I'm running 8-CURRENT on my machine and it appears that one of
the disks in my RAID5 array has taken a nose dive (BIOS recognizes
that it exists, but Intel Matrix Manager claims that the disk is an
"Offline Member"). After doing some reading it appears that it's
kaput, so I need to get a replacement disk to fix this one...
That aside, I need to determine how to rebuild the array in a
Unix environment because Intel only provides instructions for how to
use their Windows matrix manager. If anyone can point me to some links
or provide me with some pointers on how to correct this issue, I'd owe
you a lot; in fact the next time you come by Santa Cruz, CA I'll
gladly treat you to some beers or something else you might want :)...
Linux solutions (if there isn't a proper one for FreeBSD) are valid,
as long as the core data remains uncorrupted and I can do what I need
to from a LiveCD. I'm just scared to boot up OS and have it do some
irrevocable operation like fsck -y and assume parity errors are ok or
something along those lines (I don't remember if I set rc.conf to
fsck -y and I know I can change that from single-user mode, but I want
to play things conservatively if at all possible) :\...
Filesystem is UFS2 with softupdates of course.
Point proven that I need to backup my data more often :(...
TIA,
-Garrett

PS If replying on the questions@ list, please CC me as I'm not
subscribed to that list.

I'm fairly sure that Intel Matrix metadata cant be created/modified by
fbsd, only read. You should be able to do whatever rebuild operations
you need in the BIOS I believe, but that would be an offline operation,
clearly.

I may be mistaken, and the driver can handle array rebuilds for matrix.
If it can, the command you are looking for is 'atacontrol rebuild ar0',
see atacontrol(8).

Tom

Attachment: signature.asc
Description: This is a digitally signed message part

Relevant Pages

  • Re: P5LD2-Deluxe and RAID 1 rebuild...how to stop
    ... >> BIOS) from prompting them to rebuild a RAID 1 aarray t bootup. ... > When the Intel software wants to rebuild the RAID 1 array, ... > disk that is ready is marked as being an "orphan". ...
    (alt.comp.periphs.mainboard.asus)
  • Re: URGENT: Need help rebuilding iir RAID5 array with failed drive
    ... so I need to get a replacement disk to fix this one... ... I'm fairly sure that Intel Matrix metadata cant be created/modified by ... You should be able to do whatever rebuild operations ... Please DONT use FreeBSD to rebuild this RAID5 unit. ...
    (freebsd-current)
  • RAID5 Problems
    ... Earlier today I noticed that one of the drives had been removed from the array for some reason I checked out the disk and nothing was wrong with it. ... When I checked back to see how things were with the rebuild, ... The drive which I hotadded back into the array showed up as a spare instead of an active part of the array, and one of the other disks decided this was a good time to fail. ...
    (Fedora)
  • Re: Raid 1 mirroring degraded;
    ... I had no problem progressing through the NVidia Raid bios settings. ... Array Detail screen hit 'R' to rebuild which automatically returned me ... Once in Windows there was no disk building activity, ...
    (microsoft.public.windowsxp.help_and_support)
  • Re: Mechanism to safely force repair of single md stripe w/o hurting data integrity of file system
    ... Subj: RE: Mechanism to safely force repair of single md stripe w/o hurting data integrity of file system ... } balancing might result in the kernel reading the bad block from the good} disk. ... I have system with}>100TB and if I did a rebuild every time I discovered a bad block} somewhere, then a full parity repair would never complete before another} physical bad block is discovered.) ... And I would say that if you have> 100TB in a single raid5/6 that would mean you had to have at least 100 disks in that array, and most people get nervous at>8-16 disks in either raid5 or raid6 arrays, and the statistics of disks going bad, and the chance of a rebuild succeeding before another disk/block goes bad gets smaller and smaller as the number of disks increase, as you have noted you are at the point that it becomes unlikely that the rebuild will ever complete even with good disks in the array. ...
    (Linux-Kernel)