Greetings all,
I've stumbled on what seems like a really nasty problem, where after
adding a new disk, a different perfectly good disk (as of a few minutes
prior) comes to have a large number of reallocated sectors and the
mageia 'smart' software is screaming at me that the disk is about to
fail _now_ and to back it up.
Context:
mageia 1
7 sata disks.
bought a new (my first) 3TB disk, used gdisk to partition it, 'L' for
partition types, chose the last option for MBR (was this wrong?) Next
created an encrypted filesystem on this:
a) modprobe dm-mod
/usr/local/bin/cryptsetup.sh -c aes -h ripemd160 -y -b `blockdev
--getsize /dev/
sdg1` create cryptvol2 /dev/sdg1
b) mke2fs -j /dev/mapper cryptvol2
c) after completion, dmsetup remove cryptvol2
d) mount the new filesystem with;
modprobe dm-mod
/usr/local/bin/cryptsetup.sh -c aes -h ripemd160 -y -b `blockdev
--getsize /dev/
sdg1` create cryptvol2 /dev/sdg1
mount /dev/mapper/cryptvol2 /mnt/crypt2
All looked good, but there was a 'smart' warning icon about a disk about
to fail. The 3 TB disk I'd just done happened to be /dev/sdg. The disk
reporting errors was /dev/sdh, a 2TB disk.
Device: /dev/sdh
SMART status: enabled
Automatic Offline Data Collection status: enabled
ALERT: The drive is reporting that it will FAIL very soon.
Please back up as soon as possible!
I grumbled, thought it a strange coincidence that this just appeared,
took out the 'faulty' disk, bought another 3TB disk to replace it, did
it all again. The new 3TB disk as again seen as /dev/sdg. Did same
stuff again, up came the same error message about yet _another_ disk.
Having had zero reallocated sectors only a few minutes before, it now
reports 3975 reallocated sectors, the identical number to the first
'normal' disk, and same error messages. The 'faulty' disk was again the
one seen as /dev/sdg. This reallocation seems to have welded itself into
the disk; persists after reboot.
Help! I've no idea where to go next. Looks like a major bug, but in what?
