| 1 |
Hi List, |
| 2 |
|
| 3 |
A few weeks ago I created a Raid 5 set for my data partition using mdadm. |
| 4 |
|
| 5 |
This works fine, but last week a short power outage caused the server to |
| 6 |
reboot. |
| 7 |
|
| 8 |
When I came home the server was in need of maintainance because it couldn't |
| 9 |
check all his filesystems. |
| 10 |
|
| 11 |
Strange fact was that the md device wasn't start... |
| 12 |
mdadm --assemble /dev/md1 works fine, the /etc/mdadm.conf is read the the |
| 13 |
raid device is started... |
| 14 |
|
| 15 |
But why doesn't start it at boot? |
| 16 |
|
| 17 |
A piece of output from dmesg: |
| 18 |
|
| 19 |
md: Autodetecting RAID arrays. |
| 20 |
md: autorun ... |
| 21 |
md: considering hdb1 ... |
| 22 |
md: adding hdb1 ... |
| 23 |
md: created md1 |
| 24 |
md: bind<hdb1> |
| 25 |
md: running: <hdb1> |
| 26 |
raid5: device hdb1 operational as raid disk 0 |
| 27 |
raid5: not enough operational devices for md1 (2/3 failed) |
| 28 |
RAID5 conf printout: |
| 29 |
--- rd:3 wd:1 fd:2 |
| 30 |
disk 0, o:1, dev:hdb1 |
| 31 |
raid5: failed to run raid set md1 |
| 32 |
md: pers->run() failed ... |
| 33 |
md: do_md_run() returned -5 |
| 34 |
md: md1 stopped. |
| 35 |
md: unbind<hdb1> |
| 36 |
md: export_rdev(hdb1) |
| 37 |
md: ... autorun DONE. |
| 38 |
|
| 39 |
It tell's me that 2/3 devices failed, but after I assemble it.. the ouput of |
| 40 |
mdadm --detail /dev/md1 is: |
| 41 |
|
| 42 |
/dev/md1: |
| 43 |
Version : 00.90.03 |
| 44 |
Creation Time : Fri Sep 22 22:28:43 2006 |
| 45 |
Raid Level : raid5 |
| 46 |
Array Size : 625137152 (596.18 GiB 640.14 GB) |
| 47 |
Device Size : 312568576 (298.09 GiB 320.07 GB) |
| 48 |
Raid Devices : 3 |
| 49 |
Total Devices : 3 |
| 50 |
Preferred Minor : 1 |
| 51 |
Persistence : Superblock is persistent |
| 52 |
|
| 53 |
Intent Bitmap : Internal |
| 54 |
|
| 55 |
Update Time : Wed Nov 15 10:56:36 2006 |
| 56 |
State : active |
| 57 |
Active Devices : 3 |
| 58 |
Working Devices : 3 |
| 59 |
Failed Devices : 0 |
| 60 |
Spare Devices : 0 |
| 61 |
|
| 62 |
Layout : left-symmetric |
| 63 |
Chunk Size : 64K |
| 64 |
|
| 65 |
UUID : 82544aad:a2e92ea2:72ca2d55:716dada0 |
| 66 |
Events : 0.1486768 |
| 67 |
|
| 68 |
Number Major Minor RaidDevice State |
| 69 |
0 3 65 0 active sync /dev/hdb1 |
| 70 |
1 22 1 1 active sync /dev/hdc1 |
| 71 |
2 22 65 2 active sync /dev/hdd1 |
| 72 |
|
| 73 |
No failed devices! |
| 74 |
|
| 75 |
How it this possible? |
| 76 |
I rebooted aferwards, but still the same issue! :-( |
| 77 |
|
| 78 |
Anybody any clue's?! |
| 79 |
Any help is appriciated.... |
| 80 |
|
| 81 |
-- |
| 82 |
Met vriendelijke groet / With kind regards, |
| 83 |
|
| 84 |
H. van Wees |
| 85 |
--- |
| 86 |
If UNIX isn't the solution, you've got the wrong problem. |
Archives