Hello,

I have a machine with 4 IDE disks with the following raid
configuration.

cat /proc/mdstat
Personalities : [raid1] [raid5]
read_ahead 1024 sectors
md2 : active raid5 hdd5[3] hdb5[2] hdc5[1] hda5[0]
844809600 blocks level 5, 128k chunk, algorithm 2 [4/4] [UUUU]

md1 : active raid1 hda2[0] hdc2[1]
10241344 blocks [2/2] [UU]

md0 : active raid1 hda1[0] hdc1[1]
104320 blocks [2/2] [UU]

back in its healthy days. md0 is /boot, md1 is / and md2 is mount as
/home and is 831GB although usage is only 21GB.

>From the logs you can see that the raid has failed on hda5 and hdb5 and
hda2

planet kernel: raid1: Disk failure on hda2, disabling device.
planet kernel: raid5: Disk failure on hdb5, disabling device. Operation
continuing on 3 devices
planet kernel: raid5: Disk failure on hda5, disabling device. Operation
continuing on 2 devices

And if you now do a: cat /proc/mdstat

Personalities : [raid1] [raid5]
read_ahead 1024 sectors
md2 : active raid5 hdd5[3] hdb5[2](F) hdc5[1] hda5[0](F)
844809600 blocks level 5, 128k chunk, algorithm 2 [4/2] [_U_U]

md1 : active raid1 hda2[0](F) hdc2[1]
10241344 blocks [2/1] [_U]

md0 : active raid1 hda1[0](F) hdc1[1]
104320 blocks [2/1] [_U]

You see from the above that hdb[5](F) and hda5[0](F) have failed in the
raid5 set. Is this possible? 2 of 4 disks have failed in the raid 5 set
yet md2 is still mounted and readable?

The thought the definition of raid5 was: If one of the disks fail, all
data are
still intact, thanks to the parity information. If spare disks are
available, reconstruction will begin immediately after the device
failure.
If two disks fail simultaneously, all data is lost. RAID-5 can survive
one disk failure, but not two or more.

Can anyone give me any advice on what I am actually looking at with the
raid set md2 - is it recoverable or not?

This is made even more complicated by the fact that the computer is
located at the south pole and I am in Sydney though there are people at
the pole who can help I just have to make sure I give them the correct
advice.

Thanks,

Melinda