Skip to contentSkip to conversation

Raid 1 fails on every reboot?

DISCUSSION John McMonagle 4 participants 6 messages May 1, 2005 – May 2, 2005 MD RAID

lore
Share & tools
lore
#0 lore raw

I cloned a system with 2 drives and 2 raid 1 partitions.

I took a drive out of the old system.
Recreated the raids to get a new uuids.
Partitioned a new drive and added the partitions.
There were a couple other problems that I'm pretty sure was caused by a bad sata controller but think I have that under control.

So here is my problem.
Any time I reboot only the partitions on the first drive are active. If I switch drives it's always the first one that is active.

Here is my mdadm.conf
DEVICE /dev/sd*
ARRAY /dev/md0 level=raid1 num-devices=2 UUID=647381d7:fb84da16:f43e8c51:3695f234 ARRAY /dev/md1 level=raid1 num-devices=2 UUID=90b9e6d2:c78b1827:77f469ad:0f8997ed

This is all after reboot:
cat /proc/mdstat
Personalities : [raid1]
md0 : active raid1 sda1[1]

     586240 blocks [2/1] [_U]

md1 : active raid1 sda2[1]

     194771968 blocks [2/1] [_U]

mdadm -E /dev/sda1
/dev/sda1:

         Magic : a92b4efc
       Version : 00.90.00
          UUID : 647381d7:fb84da16:f43e8c51:3695f234

Creation Time : Sat Apr 30 06:02:47 2005
Raid Level : raid1
Raid Devices : 2
Total Devices : 1
Preferred Minor : 0

Update Time : Sat Apr 30 21:47:22 2005
State : clean
Active Devices : 1
Working Devices : 1
Failed Devices : 0
Spare Devices : 0
Checksum : b8e8e847 - correct
Events : 0.4509

Number Major Minor RaidDevice State
this 1 8 1 1 active sync /dev/sda1

  0     0       0        0        0      removed
  1     1       8        1        1      active sync   /dev/sda1

mdadm -E /dev/sdb1
/dev/sdb1:

         Magic : a92b4efc
       Version : 00.90.00
          UUID : 647381d7:fb84da16:f43e8c51:3695f234

Creation Time : Sat Apr 30 06:02:47 2005
Raid Level : raid1
Raid Devices : 2
Total Devices : 2
Preferred Minor : 0

Update Time : Sat Apr 30 21:26:07 2005
State : clean
Active Devices : 2
Working Devices : 2
Failed Devices : 0
Spare Devices : 0
Checksum : b8e8e1d8 - correct
Events : 0.4303

Number Major Minor RaidDevice State
this 0 8 17 0 active sync /dev/sdb1

  0     0       8       17        0      active sync   /dev/sdb1
  1     1       8        1        1      active sync   /dev/sda1

After adding /dev/sdb1 to md0:

cat /proc/mdstat
Personalities : [raid1]
md0 : active raid1 sdb1[0] sda1[1]

     586240 blocks [2/2] [UU]

md1 : active raid1 sda2[1]

     194771968 blocks [2/1] [_U]

unused devices: <none>
mdadm -E /dev/sda1
/dev/sda1:

         Magic : a92b4efc
       Version : 00.90.00
          UUID : 647381d7:fb84da16:f43e8c51:3695f234

Creation Time : Sat Apr 30 06:02:47 2005
Raid Level : raid1
Raid Devices : 2
Total Devices : 2
Preferred Minor : 0

Update Time : Sat Apr 30 21:51:11 2005
State : clean
Active Devices : 2
Working Devices : 2
Failed Devices : 0
Spare Devices : 0
Checksum : b8e8e99a - correct
Events : 0.4551

Number Major Minor RaidDevice State
this 1 8 1 1 active sync /dev/sda1

  0     0       8       17        0      active sync   /dev/sdb1
  1     1       8        1        1      active sync   /dev/sda1
mdadm -E /dev/sdb1
/dev/sdb1:
         Magic : a92b4efc
       Version : 00.90.00
          UUID : 647381d7:fb84da16:f43e8c51:3695f234

Creation Time : Sat Apr 30 06:02:47 2005
Raid Level : raid1
Raid Devices : 2
Total Devices : 2
Preferred Minor : 0

Update Time : Sat Apr 30 21:51:43 2005
State : clean
Active Devices : 2
Working Devices : 2
Failed Devices : 0
Spare Devices : 0
Checksum : b8e8e9d0 - correct
Events : 0.4555

Number Major Minor RaidDevice State
this 0 8 17 0 active sync /dev/sdb1

  0     0       8       17        0      active sync   /dev/sdb1
  1     1       8        1        1      active sync   /dev/sda1

Any Idea what is wrong?

John

#1 lore raw

well, I know it's too late for most of us in the US -

My Seagate 120 SATA system has augured in, and NO, drattit, it wasn't properly redundant. ANYWAY..

Anyone know of some place (open on Sunday, in the Philly/Princeton area) where I can get a good deal on good drives - SCSI would be great ~120Gb, PATA is 2nd choice, (I've had a lot of problems in SATA connectors [harness stiffness, I think] . so I don't want to play that fun again)? This time I'll raid them 1st. BTW the other boxes are running fine.

Thanks
b-

#2 lore raw
In reply to John McMonagle
Quoted 152 lines · John McMonagle

I cloned a system with 2 drives and 2 raid 1 partitions.

I took a drive out of the old system.
Recreated the raids to get a new uuids.
Partitioned a new drive and added the partitions.
There were a couple other problems that I'm pretty sure was caused by a bad sata controller but think I have that under control.

So here is my problem.
Any time I reboot only the partitions on the first drive are active. If I switch drives it's always the first one that is active.

Here is my mdadm.conf
DEVICE /dev/sd*
ARRAY /dev/md0 level=raid1 num-devices=2 UUID=647381d7:fb84da16:f43e8c51:3695f234 ARRAY /dev/md1 level=raid1 num-devices=2 UUID=90b9e6d2:c78b1827:77f469ad:0f8997ed

This is all after reboot:
cat /proc/mdstat
Personalities : [raid1]
md0 : active raid1 sda1[1]

     586240 blocks [2/1] [_U]

md1 : active raid1 sda2[1]

     194771968 blocks [2/1] [_U]

mdadm -E /dev/sda1
/dev/sda1:

         Magic : a92b4efc
       Version : 00.90.00
          UUID : 647381d7:fb84da16:f43e8c51:3695f234

Creation Time : Sat Apr 30 06:02:47 2005
Raid Level : raid1
Raid Devices : 2
Total Devices : 1
Preferred Minor : 0

Update Time : Sat Apr 30 21:47:22 2005
State : clean
Active Devices : 1
Working Devices : 1
Failed Devices : 0
Spare Devices : 0
Checksum : b8e8e847 - correct
Events : 0.4509

Number Major Minor RaidDevice State
this 1 8 1 1 active sync /dev/sda1

  0     0       0        0        0      removed
  1     1       8        1        1      active sync   /dev/sda1

mdadm -E /dev/sdb1
/dev/sdb1:

         Magic : a92b4efc
       Version : 00.90.00
          UUID : 647381d7:fb84da16:f43e8c51:3695f234

Creation Time : Sat Apr 30 06:02:47 2005
Raid Level : raid1
Raid Devices : 2
Total Devices : 2
Preferred Minor : 0

Update Time : Sat Apr 30 21:26:07 2005
State : clean
Active Devices : 2
Working Devices : 2
Failed Devices : 0
Spare Devices : 0
Checksum : b8e8e1d8 - correct
Events : 0.4303

Number Major Minor RaidDevice State
this 0 8 17 0 active sync /dev/sdb1

  0     0       8       17        0      active sync   /dev/sdb1
  1     1       8        1        1      active sync   /dev/sda1

After adding /dev/sdb1 to md0:

cat /proc/mdstat
Personalities : [raid1]
md0 : active raid1 sdb1[0] sda1[1]

     586240 blocks [2/2] [UU]

md1 : active raid1 sda2[1]

     194771968 blocks [2/1] [_U]

unused devices: <none>
mdadm -E /dev/sda1
/dev/sda1:

         Magic : a92b4efc
       Version : 00.90.00
          UUID : 647381d7:fb84da16:f43e8c51:3695f234

Creation Time : Sat Apr 30 06:02:47 2005
Raid Level : raid1
Raid Devices : 2
Total Devices : 2
Preferred Minor : 0

Update Time : Sat Apr 30 21:51:11 2005
State : clean
Active Devices : 2
Working Devices : 2
Failed Devices : 0
Spare Devices : 0
Checksum : b8e8e99a - correct
Events : 0.4551

Number Major Minor RaidDevice State
this 1 8 1 1 active sync /dev/sda1

  0     0       8       17        0      active sync   /dev/sdb1
  1     1       8        1        1      active sync   /dev/sda1
mdadm -E /dev/sdb1
/dev/sdb1:
         Magic : a92b4efc
       Version : 00.90.00
          UUID : 647381d7:fb84da16:f43e8c51:3695f234

Creation Time : Sat Apr 30 06:02:47 2005
Raid Level : raid1
Raid Devices : 2
Total Devices : 2
Preferred Minor : 0

Update Time : Sat Apr 30 21:51:43 2005
State : clean
Active Devices : 2
Working Devices : 2
Failed Devices : 0
Spare Devices : 0
Checksum : b8e8e9d0 - correct
Events : 0.4555

Number Major Minor RaidDevice State
this 0 8 17 0 active sync /dev/sdb1

  0     0       8       17        0      active sync   /dev/sdb1
  1     1       8        1        1      active sync   /dev/sda1

Any Idea what is wrong?

John

Quoted 4 lines

-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

what does fdisk -l says?

Signature · 6 lines
Don't smoke the next cigarette.  Repeat.

-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
#3 lore raw
Quoted 146 lines · Laurent CARON
Quoted 144 lines · John McMonagle

I cloned a system with 2 drives and 2 raid 1 partitions.

I took a drive out of the old system.
Recreated the raids to get a new uuids.
Partitioned a new drive and added the partitions.
There were a couple other problems that I'm pretty sure was caused by a bad sata controller but think I have that under control.

So here is my problem.
Any time I reboot only the partitions on the first drive are active. If I switch drives it's always the first one that is active.

Here is my mdadm.conf
DEVICE /dev/sd*
ARRAY /dev/md0 level=raid1 num-devices=2 UUID=647381d7:fb84da16:f43e8c51:3695f234 ARRAY /dev/md1 level=raid1 num-devices=2 UUID=90b9e6d2:c78b1827:77f469ad:0f8997ed

This is all after reboot:
cat /proc/mdstat
Personalities : [raid1]
md0 : active raid1 sda1[1]

     586240 blocks [2/1] [_U]

md1 : active raid1 sda2[1]

     194771968 blocks [2/1] [_U]

mdadm -E /dev/sda1
/dev/sda1:

         Magic : a92b4efc
       Version : 00.90.00
          UUID : 647381d7:fb84da16:f43e8c51:3695f234

Creation Time : Sat Apr 30 06:02:47 2005
Raid Level : raid1
Raid Devices : 2
Total Devices : 1
Preferred Minor : 0

Update Time : Sat Apr 30 21:47:22 2005
State : clean
Active Devices : 1
Working Devices : 1
Failed Devices : 0
Spare Devices : 0
Checksum : b8e8e847 - correct
Events : 0.4509

Number Major Minor RaidDevice State
this 1 8 1 1 active sync /dev/sda1

  0     0       0        0        0      removed
  1     1       8        1        1      active sync   /dev/sda1

mdadm -E /dev/sdb1
/dev/sdb1:

         Magic : a92b4efc
       Version : 00.90.00
          UUID : 647381d7:fb84da16:f43e8c51:3695f234

Creation Time : Sat Apr 30 06:02:47 2005
Raid Level : raid1
Raid Devices : 2
Total Devices : 2
Preferred Minor : 0

Update Time : Sat Apr 30 21:26:07 2005
State : clean
Active Devices : 2
Working Devices : 2
Failed Devices : 0
Spare Devices : 0
Checksum : b8e8e1d8 - correct
Events : 0.4303

Number Major Minor RaidDevice State
this 0 8 17 0 active sync /dev/sdb1

  0     0       8       17        0      active sync   /dev/sdb1
  1     1       8        1        1      active sync   /dev/sda1

After adding /dev/sdb1 to md0:

cat /proc/mdstat
Personalities : [raid1]
md0 : active raid1 sdb1[0] sda1[1]

     586240 blocks [2/2] [UU]

md1 : active raid1 sda2[1]

     194771968 blocks [2/1] [_U]

unused devices: <none>
mdadm -E /dev/sda1
/dev/sda1:

         Magic : a92b4efc
       Version : 00.90.00
          UUID : 647381d7:fb84da16:f43e8c51:3695f234

Creation Time : Sat Apr 30 06:02:47 2005
Raid Level : raid1
Raid Devices : 2
Total Devices : 2
Preferred Minor : 0

Update Time : Sat Apr 30 21:51:11 2005
State : clean
Active Devices : 2
Working Devices : 2
Failed Devices : 0
Spare Devices : 0
Checksum : b8e8e99a - correct
Events : 0.4551

Number Major Minor RaidDevice State
this 1 8 1 1 active sync /dev/sda1

  0     0       8       17        0      active sync   /dev/sdb1
  1     1       8        1        1      active sync   /dev/sda1
mdadm -E /dev/sdb1
/dev/sdb1:
         Magic : a92b4efc
       Version : 00.90.00
          UUID : 647381d7:fb84da16:f43e8c51:3695f234

Creation Time : Sat Apr 30 06:02:47 2005
Raid Level : raid1
Raid Devices : 2
Total Devices : 2
Preferred Minor : 0

Update Time : Sat Apr 30 21:51:43 2005
State : clean
Active Devices : 2
Working Devices : 2
Failed Devices : 0
Spare Devices : 0
Checksum : b8e8e9d0 - correct
Events : 0.4555

Number Major Minor RaidDevice State
this 0 8 17 0 active sync /dev/sdb1

  0     0       8       17        0      active sync   /dev/sdb1
  1     1       8        1        1      active sync   /dev/sda1
Quoted 8 lines
Quoted 5 lines

Any Idea what is wrong?

John

what does fdisk -l says?

neebackup:~# fdisk -l

Disk /dev/sda: 200.0 GB, 200049647616 bytes
255 heads, 63 sectors/track, 24321 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

Device Boot Start End Blocks Id System

/dev/sda1   *           1          73      586341   fd  Linux raid autodetect
/dev/sda2              74       24321   194772060   fd  Linux raid autodetect

Disk /dev/sdb: 200.0 GB, 200049647616 bytes
255 heads, 63 sectors/track, 24321 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

Device Boot Start End Blocks Id System

/dev/sdb1   *           1          73      586341   fd  Linux raid autodetect
/dev/sdb2              74       24321   194772060   fd  Linux raid autodetect

Also a lttle more info.

It's debian sarge.

It's a SATA150 TX4 sata controler that had been in use for months.

neebackup:~# mdadm -D /dev/md0
/dev/md0:
       Version : 00.90.01
 Creation Time : Sat Apr 30 06:02:47 2005
    Raid Level : raid1
    Array Size : 586240 (572.50 MiB 600.31 MB)
   Device Size : 586240 (572.50 MiB 600.31 MB)
  Raid Devices : 2
 Total Devices : 1
Preferred Minor : 0
   Persistence : Superblock is persistent

Update Time : Sun May 1 08:40:40 2005
State : clean, degraded
Active Devices : 1
Working Devices : 1
Failed Devices : 0
Spare Devices : 0

          UUID : 647381d7:fb84da16:f43e8c51:3695f234
        Events : 0.4875

Number Major Minor RaidDevice State

      0       8        1        0      active sync   /dev/sda1
      1       0        0        -      removed

There are no disk error messages.

I booted with a fedora 3 rescue cd and md0 started properly so I assume it's something in the startup.

I rebuilt the initrd a couple times. Is there anything in it that could cause this problem?

John

-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#4 lore raw

See the cause in the start up script in the initrd:

mdadm -A /devfs/md/1 -R -u 90b9e6d2:c78b1827:77f469ad:0f8997ed /dev/sda2 mkdir /devfs/vg1
mount_tmpfs /var
if [ -f /etc/lvm/lvm.conf ]; then
cat /etc/lvm/lvm.conf > /var/lvm.conf
fi
mount_tmpfs /etc/lvm
if [ -f /var/lvm.conf ]; then
cat /var/lvm.conf > /etc/lvm/lvm.conf
fi
mount -nt devfs devfs /dev
vgchange -a y vg1
umount /dev
umount -n /var
umount -n /etc/lvm
ROOT=/dev/md0
mdadm -A /devfs/md/0 -R -u 647381d7:fb84da16:f43e8c51:3695f234 /dev/sda1

This is debian sarge with 2.6.11 kernel.

Can make it work by adding the partitions to md0 and md1 and then rebuild the initrd.

While this gets me running it doesn't seem ideal.
There is a good chance of changing devices without rebuilding the initrd.

Is it possible to do the mdadm -A with out specifying the devices such as? mdadm -A /devfs/md/0 -R -u 647381d7:fb84da16:f43e8c51:3695f234

If so it's something to bug the debian folks about.

Any other way it should be done?

Thanks

John

Quoted 211 lines · John McMonagle
Quoted 146 lines · Laurent CARON
Quoted 144 lines · John McMonagle

I cloned a system with 2 drives and 2 raid 1 partitions.

I took a drive out of the old system.
Recreated the raids to get a new uuids.
Partitioned a new drive and added the partitions.
There were a couple other problems that I'm pretty sure was caused by a bad sata controller but think I have that under control.

So here is my problem.
Any time I reboot only the partitions on the first drive are active. If I switch drives it's always the first one that is active.

Here is my mdadm.conf
DEVICE /dev/sd*
ARRAY /dev/md0 level=raid1 num-devices=2 UUID=647381d7:fb84da16:f43e8c51:3695f234 ARRAY /dev/md1 level=raid1 num-devices=2 UUID=90b9e6d2:c78b1827:77f469ad:0f8997ed

This is all after reboot:
cat /proc/mdstat
Personalities : [raid1]
md0 : active raid1 sda1[1]

     586240 blocks [2/1] [_U]

md1 : active raid1 sda2[1]

     194771968 blocks [2/1] [_U]

mdadm -E /dev/sda1
/dev/sda1:

         Magic : a92b4efc
       Version : 00.90.00
          UUID : 647381d7:fb84da16:f43e8c51:3695f234

Creation Time : Sat Apr 30 06:02:47 2005
Raid Level : raid1
Raid Devices : 2
Total Devices : 1
Preferred Minor : 0

Update Time : Sat Apr 30 21:47:22 2005
State : clean
Active Devices : 1
Working Devices : 1
Failed Devices : 0
Spare Devices : 0
Checksum : b8e8e847 - correct
Events : 0.4509

Number Major Minor RaidDevice State
this 1 8 1 1 active sync /dev/sda1

  0     0       0        0        0      removed
  1     1       8        1        1      active sync   /dev/sda1

mdadm -E /dev/sdb1
/dev/sdb1:

         Magic : a92b4efc
       Version : 00.90.00
          UUID : 647381d7:fb84da16:f43e8c51:3695f234

Creation Time : Sat Apr 30 06:02:47 2005
Raid Level : raid1
Raid Devices : 2
Total Devices : 2
Preferred Minor : 0

Update Time : Sat Apr 30 21:26:07 2005
State : clean
Active Devices : 2
Working Devices : 2
Failed Devices : 0
Spare Devices : 0
Checksum : b8e8e1d8 - correct
Events : 0.4303

Number Major Minor RaidDevice State
this 0 8 17 0 active sync /dev/sdb1

  0     0       8       17        0      active sync   /dev/sdb1
  1     1       8        1        1      active sync   /dev/sda1

After adding /dev/sdb1 to md0:

cat /proc/mdstat
Personalities : [raid1]
md0 : active raid1 sdb1[0] sda1[1]

     586240 blocks [2/2] [UU]

md1 : active raid1 sda2[1]

     194771968 blocks [2/1] [_U]

unused devices: <none>
mdadm -E /dev/sda1
/dev/sda1:

         Magic : a92b4efc
       Version : 00.90.00
          UUID : 647381d7:fb84da16:f43e8c51:3695f234

Creation Time : Sat Apr 30 06:02:47 2005
Raid Level : raid1
Raid Devices : 2
Total Devices : 2
Preferred Minor : 0

Update Time : Sat Apr 30 21:51:11 2005
State : clean
Active Devices : 2
Working Devices : 2
Failed Devices : 0
Spare Devices : 0
Checksum : b8e8e99a - correct
Events : 0.4551

Number Major Minor RaidDevice State
this 1 8 1 1 active sync /dev/sda1

  0     0       8       17        0      active sync   /dev/sdb1
  1     1       8        1        1      active sync   /dev/sda1
mdadm -E /dev/sdb1
/dev/sdb1:
         Magic : a92b4efc
       Version : 00.90.00
          UUID : 647381d7:fb84da16:f43e8c51:3695f234

Creation Time : Sat Apr 30 06:02:47 2005
Raid Level : raid1
Raid Devices : 2
Total Devices : 2
Preferred Minor : 0

Update Time : Sat Apr 30 21:51:43 2005
State : clean
Active Devices : 2
Working Devices : 2
Failed Devices : 0
Spare Devices : 0
Checksum : b8e8e9d0 - correct
Events : 0.4555

Number Major Minor RaidDevice State
this 0 8 17 0 active sync /dev/sdb1

  0     0       8       17        0      active sync   /dev/sdb1
  1     1       8        1        1      active sync   /dev/sda1
Quoted 8 lines
Quoted 5 lines

Any Idea what is wrong?

John

what does fdisk -l says?

neebackup:~# fdisk -l

Disk /dev/sda: 200.0 GB, 200049647616 bytes
255 heads, 63 sectors/track, 24321 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

Device Boot Start End Blocks Id System

/dev/sda1   *           1          73      586341   fd  Linux raid autodetect
/dev/sda2              74       24321   194772060   fd  Linux raid autodetect

Disk /dev/sdb: 200.0 GB, 200049647616 bytes
255 heads, 63 sectors/track, 24321 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

Device Boot Start End Blocks Id System

/dev/sdb1   *           1          73      586341   fd  Linux raid autodetect
/dev/sdb2              74       24321   194772060   fd  Linux raid autodetect

Also a lttle more info.

It's debian sarge.

It's a SATA150 TX4 sata controler that had been in use for months.

neebackup:~# mdadm -D /dev/md0
/dev/md0:
       Version : 00.90.01
 Creation Time : Sat Apr 30 06:02:47 2005
    Raid Level : raid1
    Array Size : 586240 (572.50 MiB 600.31 MB)
   Device Size : 586240 (572.50 MiB 600.31 MB)
  Raid Devices : 2
 Total Devices : 1
Preferred Minor : 0
   Persistence : Superblock is persistent

Update Time : Sun May 1 08:40:40 2005
State : clean, degraded
Active Devices : 1
Working Devices : 1
Failed Devices : 0
Spare Devices : 0

          UUID : 647381d7:fb84da16:f43e8c51:3695f234
        Events : 0.4875

Number Major Minor RaidDevice State

      0       8        1        0      active sync   /dev/sda1
      1       0        0        -      removed

There are no disk error messages.

I booted with a fedora 3 rescue cd and md0 started properly so I assume it's something in the startup.

I rebuilt the initrd a couple times. Is there anything in it that could cause this problem?

Quoted 7 lines

John

-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#5 lore raw
In reply to John McMonagle

----- Original Message ----- From: "John McMonagle"

Quoted 3 lines

So here is my problem.
Any time I reboot only the partitions on the first drive are active. If I switch drives it's always the first one that is active.

I'm not sure about scsi drives, but for RAID-1 IDE drives I find that I have to add a -h argument to the halt script.

The last lines of:
/etc/init.d/halt

are:

HALTARGS="-i -d -h"
[ -f /poweroff -o ! -f /halt ] && HALTARGS="$HALTARGS -p"

exec $command $HALTARGS

What I have found is that by adding a -h (man halt: Put all harddrives on the system in standby mode just before halt or poweroff), this somehow flushes the drives and I don't have rebuild problems upon reboot. However, this works for FC 1 systems and I do not know if this is still neccessary for other distros.

-Eric Wood