You are not logged in.
Hi everyone,
Last week I updated one of my computers, which has not been updated for a year or so (it was offline too).
Since the update and reboot, I can't seem to access a mdadm array which was used for storage.
My kernel is the stock archlinux one:
[13:35][root@plop][~]# uname -a
Linux plop 3.17.6-1-ARCH #1 SMP PREEMPT Sun Dec 7 23:43:32 UTC 2014 x86_64 GNU/Linux
The md and raid modules seem to be correcly loaded:
[13:41][root@plop][~]# lsmod | grep -i '\(md\|raid\)'
raid456 81478 0
async_raid6_recov 16582 1 raid456
async_memcpy 12350 2 raid456,async_raid6_recov
async_pq 12476 2 raid456,async_raid6_recov
async_xor 12385 3 async_pq,raid456,async_raid6_recov
async_tx 12566 5 async_pq,raid456,async_xor,async_memcpy,async_raid6_recov
raid6_pq 99334 2 async_pq,async_raid6_recov
raid1 34552 0
md_mod 115864 2 raid456,raid1
kvm_amd 59131 0
kvm 421519 1 kvm_amd
edac_mce_amd 21166 0
The array is found in the disk superblocks (it is located on the whole sdb sdc and sdd drives):
[13:41][root@salon][~]# mdadm -E --scan
ARRAY /dev/md/666 metadata=1.2 UUID=d917d21c:be76bfcc:b8261743:82dd8cd0 name=plop:666
My /etc/mdadm.conf file is:
[13:48][root@plop][~]# cat /etc/mdadm.conf | grep -v '#'
DEVICE /dev/sdb /dev/sdc /dev/sdd
ARRAY /dev/md/666 metadata=1.2 UUID=d917d21c:be76bfcc:b8261743:82dd8cd0 name=plop:666
However my device is inaccessible in /dev. There is no /dev/md* file or directory whatsoever.
The array is not mentionned in /proc/mdstat:
[13:54][root@plop][~]# cat /proc/mdstat
Personalities : [raid6] [raid5] [raid4]
unused devices: <none>
And when I try to force assemble it:
[13:58][root@plop][~]# mdadm -A --scan
mdadm: unexpected failure opening /dev/md666
[13:57][root@plop][~]# mdadm -A /dev/md/666
mdadm: unexpected failure opening /dev/md666
Any idea anyone ?
Offline
Maybe the output of mdadm -E /dev/sd[bcd] would be informative.
But whether the Constitution really be one thing, or another, this much is certain - that it has either authorized such a government as we have had, or has been powerless to prevent it. In either case, it is unfit to exist.
-Lysander Spooner
Offline
Why are you posting the output from two (apparently) different hosts?
Offline
@alphaniner: here is the full information for the disks
[00:33][root@plop][~]# mdadm -E /dev/sd[bcd]
/dev/sdb
Magic : a92b4efc
Version : 1.2
Feature Map : 0x0
Array UUID : d917d21c:be76bfcc:b8261743:82dd8cd0
Name : plop:666 (local to host plop)
Creation Time : Sat Dec 15 10:37:16 2012
Raid Level : raid5
Raid Devices : 3
Avail Dev Size : 5860271024 (2794.40 GiB 3000.46 GB)
Array Size : 5860270080 (5588.79 GiB 6000.92 GB)
Used Dev Size : 5860270080 (2794.39 GiB 3000.46 GB)
Data Offset : 262144 sectors
Super Offset : 8 sectors
Unused Space : before=262064 sectors, after=944 sectors
State : clean
Device UUID : e84c1b10:8c9d30c6:35b2f255:9de0581c
Update Time : Fri Oct 10 00:26:50 2014
Checksum : f402a6cd - correct
Events : 70435
Layout : left-symmetric
Chunk Size : 512K
Device Role : Active device 0
Array State : AAA ('A' == active, '.' == missing, 'R' == replacing)
/dev/sdc:
Magic : a92b4efc
Version : 1.2
Feature Map : 0x0
Array UUID : d917d21c:be76bfcc:b8261743:82dd8cd0
Name : plop:666 (local to host plop)
Creation Time : Sat Dec 15 10:37:16 2012
Raid Level : raid5
Raid Devices : 3
Avail Dev Size : 5860271024 (2794.40 GiB 3000.46 GB)
Array Size : 5860270080 (5588.79 GiB 6000.92 GB)
Used Dev Size : 5860270080 (2794.39 GiB 3000.46 GB)
Data Offset : 262144 sectors
Super Offset : 8 sectors
Unused Space : before=262064 sectors, after=944 sectors
State : clean
Device UUID : 7aa650b7:6689efb9:d37dbd39:38183779
Update Time : Fri Oct 10 00:26:50 2014
Checksum : cf9fef74 - correct
Events : 70435
Layout : left-symmetric
Chunk Size : 512K
Device Role : Active device 1
Array State : AAA ('A' == active, '.' == missing, 'R' == replacing)
/dev/sdd:
Magic : a92b4efc
Version : 1.2
Feature Map : 0x0
Array UUID : d917d21c:be76bfcc:b8261743:82dd8cd0
Name : plop:666 (local to host plop)
Creation Time : Sat Dec 15 10:37:16 2012
Raid Level : raid5
Raid Devices : 3
Avail Dev Size : 5860271024 (2794.40 GiB 3000.46 GB)
Array Size : 5860270080 (5588.79 GiB 6000.92 GB)
Used Dev Size : 5860270080 (2794.39 GiB 3000.46 GB)
Data Offset : 262144 sectors
Super Offset : 8 sectors
Unused Space : before=262064 sectors, after=944 sectors
State : clean
Device UUID : 56f3c5bb:686073e2:d0677a15:7ee63488
Update Time : Fri Oct 10 00:26:50 2014
Checksum : e753cb97 - correct
Events : 70435
Layout : left-symmetric
Chunk Size : 512K
Device Role : Active device 2
Array State : AAA ('A' == active, '.' == missing, 'R' == replacing)
@jasonwryan: my bad, it all comes from the same host, just that I ineffectively tried to camouflage the real hostname when posting... Epic fail is the term.
Offline
First, use blkid to make sure that the type of all the RAID member partitions is set to "linux_raid_member".
Next, did you try to assemble the RAID?
mdadm --assemble --scan -v
Finally, it sounds like your RAID superblock has become corrupted? Try following the instructions here to rebuild your RAID superblock.
Offline
Hi,
Sorry for the lag. It was vacation time for me .
Here are the results you asked for.
[21:00][root@plop][~]# blkid
/dev/sdc: UUID="d917d21c-be76-bfcc-b826-174382dd8cd0" UUID_SUB="7aa650b7-6689-efb9-d37d-bd3938183779" LABEL="plop:666" TYPE="linux_raid_member"
/dev/sdb: UUID="d917d21c-be76-bfcc-b826-174382dd8cd0" UUID_SUB="e84c1b10-8c9d-30c6-35b2-f2559de0581c" LABEL="plop:666" TYPE="linux_raid_member"
/dev/sda1: UUID="4274e216-94f4-4830-8120-aa6451318e90" TYPE="ext2" PARTUUID="2d5509d2-01"
/dev/sda2: UUID="92339f46-71ba-41de-abe5-70740d7554c9" TYPE="ext4" PARTUUID="2d5509d2-02"
/dev/sda3: UUID="e568aecf-5059-4f9d-874f-53437c72b822" TYPE="ext4" PARTUUID="2d5509d2-03"
/dev/sda4: UUID="f8791d99-e21a-403a-81b6-255215e8df32" TYPE="swap" PARTUUID="2d5509d2-04"
/dev/sdd: UUID="d917d21c-be76-bfcc-b826-174382dd8cd0" UUID_SUB="56f3c5bb-6860-73e2-d067-7a157ee63488" LABEL="plop:666" TYPE="linux_raid_member"
[21:00][root@plop][~]# mdadm --assemble --scan -v
mdadm: looking for devices for /dev/md/666
mdadm: unexpected failure opening /dev/md666
It is not the first time such a thing happens to me. Last time, the course I took was to:
- assemble the array using a live cd (in that configuration the array was successfully assembled)
- remove a disk from it
- format that disk
- copy the data from the degraded arary
- reboot to the current system
- build a new array from the 2 disks of the old array
- copy the data to the new array
- wipe again the lone disk
- and then re-add it to the new array.
It worked, but it was cumbersome, and far too risky to my taste.
Offline
I don't see anything in the 'mdadm -E /dev/sdX' output that indicates any problems, but I'm not an expert or anything. Maybe try some live cds from between the time it was working to present to get an idea of when (eg. kernel & mdadm version) it stopped working. You might also consider the linux-raid mailing list for more help.
But whether the Constitution really be one thing, or another, this much is certain - that it has either authorized such a government as we have had, or has been powerless to prevent it. In either case, it is unfit to exist.
-Lysander Spooner
Offline