| 1 |
I also sent this one to the linux-raid-ml, I am in trouble: |
| 2 |
|
| 3 |
Sorry for maybe FAQing, I am in emergency mode: |
| 4 |
|
| 5 |
customer server, RAID5 + hotspare, 4 drives ... |
| 6 |
|
| 7 |
gentoo Linux version 2.6.25-gentoo-r7 |
| 8 |
|
| 9 |
mdadm 2.6.4-r1 here |
| 10 |
|
| 11 |
- |
| 12 |
|
| 13 |
one of the 4 drives showed massive errors in dmesg, /dev/sdc |
| 14 |
SMART-errors etc. |
| 15 |
bought new drive and wanted to swap today. |
| 16 |
|
| 17 |
# cat /proc/mdstat |
| 18 |
Personalities : [raid0] [raid1] [raid6] [raid5] [raid4] |
| 19 |
md1 : active raid1 sdb1[1] sda1[0] |
| 20 |
104320 blocks [2/2] [UU] |
| 21 |
|
| 22 |
md3 : active raid5 sdb3[1] sda3[0] |
| 23 |
19550976 blocks level 5, 64k chunk, algorithm 2 [3/2] [UU_] |
| 24 |
|
| 25 |
md4 : inactive sdb4[1](S) sdd4[3](S) sdc4[2](S) sda4[0](S) |
| 26 |
583641088 blocks |
| 27 |
|
| 28 |
|
| 29 |
|
| 30 |
- |
| 31 |
|
| 32 |
|
| 33 |
I did: |
| 34 |
|
| 35 |
mdadm /dev/md3 --fail /dev/sdc3 |
| 36 |
|
| 37 |
went OK |
| 38 |
|
| 39 |
mdadm /dev/md4 --remove /dev/sdc3 |
| 40 |
|
| 41 |
OK as well, raid md3 rebuilt |
| 42 |
|
| 43 |
- |
| 44 |
|
| 45 |
With md4 I was too aggressive maybe: |
| 46 |
|
| 47 |
mdadm /dev/md4 --fail /dev/sdc4 --remove /dev/sdc4 |
| 48 |
|
| 49 |
this rendered md4 unusable, even after a reboot it can't be reassambled. |
| 50 |
|
| 51 |
This is bad, to say the least. |
| 52 |
|
| 53 |
md4 : inactive sdb4[1](S) sdd4[3](S) sdc4[2](S) sda4[0](S) |
| 54 |
583641088 blocks |
| 55 |
|
| 56 |
|
| 57 |
What to try? |
| 58 |
|
| 59 |
This is a crucial server and I feel a lot of pressure. |
| 60 |
Rebuilding that raid would mean a lot of restore-work etc. |
| 61 |
So I would really appreciate a goo advice here. |
| 62 |
|
| 63 |
THANKS! |
| 64 |
|
| 65 |
Stefan |
Archives