Hello,
We're testing lvm over md (RAID 1) for a database application. Part of our
tests include simulating failure. So, we pulled the power on one of the
drives, and the kernel reported...
hdd: status error: status=0x00 { }
hdd: drive not ready for command
...
ide1: reset timed-out, status=0x80
hdd: lost interrupt
hdd: lost interrupt
hdd: lost interrupt
...
However, /proc/mdstat still reports:
Personalities : [linear] [raid0] [raid1] [raid5]
read_ahead 1024 sectors
md1 : active raid1 hdd[1] hdb[0]
19925760 blocks [2/2] [UU]
md0 : active raid1 hdc[1] hda[0]
19925760 blocks [2/2] [UU]
It appears the md driver just didn't bother to notice -- and now all writes
to the filesystem are blocked.
In case it matters, here's our lvm config:
--- Volume group ---
VG Name data0_vg
VG Access read/write
VG Status available/resizable
VG # 0
MAX LV 256
Cur LV 1
Open LV 1
MAX LV Size 255.99 GB
Max PV 256
Cur PV 2
Act PV 2
VG Size 38 GB
PE Size 4 MB
Total PE 9728
Alloc PE / Size 7680 / 30 GB
Free PE / Size 2048 / 8 GB
VG UUID 6XH7Nq-JL9u-cbjw-rUoP-Kx9o-hlZN-LPhdmj
--- Logical volume ---
LV Name /dev/data0_vg/data0_lv
VG Name data0_vg
LV Write Access read/write
LV Status available
LV # 1
# open 1
LV Size 30 GB
Current LE 7680
Allocated LE 7680
Stripes 2
Stripe size (KByte) 16
Allocation next free
Read ahead sectors 120
Block device 58:0
--- Physical volumes ---
PV Name (#) /dev/md0 (1)
PV Status available / allocatable
Total PE / Free PE 4864 / 1024
PV Name (#) /dev/md1 (2)
PV Status available / allocatable
Total PE / Free PE 4864 / 1024
Please CC: me on replies. Thanks.
--
Michael S. Fischer / michael at auctionwatch.com /
http://www.auctionwatch.com
Systems Engineer, AuctionWatch.com Inc. / Phone: +1 650 808 5842
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
