mdadm - dd'ed a partition. "no superblock detected"

clvn · 2011-02-20 20:59:45

(This is kind of regarding an earlier mdadm problem, but i should not be necessary to read about my previous problem. Anyway, here is my previous post)

I'm trying to repair a kind of messed up mdadm raid5 array. There are 5 out of 6 drives running, but 1 of the 5 disks (/dev/sde) got problems with some sectors that cannot be written to.

Buffer I/O error on device sde1, logical block 983126736
Buffer I/O error on device sde1, logical block 983126737
Buffer I/O error on device sde1, logical block 983126738
Buffer I/O error on device sde1, logical block 983126739
Buffer I/O error on device sde1, logical block 983126740
Buffer I/O error on device sde1, logical block 983126741
Buffer I/O error on device sde1, logical block 983126742
Buffer I/O error on device sde1, logical block 983126743
Buffer I/O error on device sde1, logical block 983126744
Buffer I/O error on device sde1, logical block 983126745

The first step towards getting this raid up and running with 6/6 healthy drives would be to make all of the 5 drives containing data to be writeable. To do that i would have to copy the content of /dev/sde1 to /dev/sda1. /dev/sda1 is a new partition with exactly the same amount of sectors.

sudo dd if=/dev/sde1 of=/dev/sda1 bs=4M conv=noerror

It ran successfully. The data at /dev/sda1 should be identical to /dev/sde1

Current partition layout of the involved disks:

#  clvn@enigma: sudo fdisk -l /dev/sd{a,e}

Disk /dev/sda: 1500.3 GB, 1500301910016 bytes
255 heads, 63 sectors/track, 182401 cylinders, total 2930277168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x000a3168

   Device Boot          Start           End     Blocks   Id  System
/dev/sda1               2048  2930274049  1465136001   fd  Linux raid autodetect

Disk /dev/sde: 1500.3 GB, 1500301910016 bytes
255 heads, 63 sectors/track, 182401 cylinders, total 2930277168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x7b48ff68

   Device Boot          Start           End     Blocks   Id  System
/dev/sde1               63  2930272064  1465136001   fd  Linux raid autodetect

(identical number of sectors, only different starting points.)

#  clvn@enigma: sudo mdadm --examine --verbose /dev/sde1
/dev/sde1:
        Magic : a92b4efc
        Version : 0.90.00
                UUID : e88a3b09:b4cd0d6e:dc342cdc:19e2ab4a
  Creation Time : Fri Oct  2 03:43:23 2009
        Raid Level : raid5
  Used Dev Size : 1465135936 (1397.26 GiB 1500.30 GB)
        Array Size : 7325679680 (6986.31 GiB 7501.50 GB)
   Raid Devices : 6
  Total Devices : 6
Preferred Minor : 0

        Update Time : Sun Feb 20 10:27:02 2011
        State : clean
 Active Devices : 5
Working Devices : 6
 Failed Devices : 1
  Spare Devices : 1
        Checksum : 2c49d439 - correct
                Events : 1735369

                Layout : left-symmetric
        Chunk Size : 64K

        Number   Major   Minor   RaidDevice State
this    4       8       65      4       active sync   /dev/sde1

   0    0       8       17      0       active sync   /dev/sdb1
   1    1       0       0       1       faulty removed
   2    2       8       97      2       active sync   /dev/sdg1
   3    3       8       49      3       active sync   /dev/sdd1
   4    4       8       65      4       active sync   /dev/sde1
   5    5       8       33      5       active sync   /dev/sdc1
   6    6       8       81      6       spare   /dev/sdf1

#  clvn@enigma: sudo mdadm --examine --verbose /dev/sda1
mdadm: No md superblock detected on /dev/sda1.

This is what happens when i try to force the raid up with the new partition containing the same data as the damaged partition:

#  clvn@enigma: sudo mdadm --assemble -v /dev/md0 --force /dev/sdc1 /dev/sdd1 /dev/sdg1 /dev/sdb1 /dev/sda1
mdadm: looking for devices for /dev/md0
mdadm: no RAID superblock on /dev/sda1
mdadm: /dev/sda1 has no superblock - assembly aborted

So.. why does not /dev/sda1 contain the same superblock as /dev/sde1? Is it possible to create a new superblock based on the superblock on the damaged disk?

(If anything in this post is hard to understand, please tell me so that i can clarify). Thanks a lot, everyone!

clvn · 2011-02-22 01:13:13

Does "Number" show partitions order in an array?

mdadm --examine --verbose /dev/sde1
[...]
        Number   Major   Minor   RaidDevice State
this    4       8       65      4       active sync   /dev/sde1

   0    0       8       17      0       active sync   /dev/sdb1
   1    1       0       0       1       faulty removed
   2    2       8       97      2       active sync   /dev/sdg1
   3    3       8       49      3       active sync   /dev/sdd1
   4    4       8       65      4       active sync   /dev/sde1
   5    5       8       33      5       active sync   /dev/sdc1
   6    6       8       81      6       spare   /dev/sdf1

And could this information be used to force creation of superblocks when combined with create-mode?

(Keep in mind that /dev/sda1 = dd'ed /dev/sde1)

mdadm --create /dev/md0 --verbose --level=5 --raid-devices=6 /dev/sdb1 missing /dev/sdg1 /dev/sdd1 /dev/sda1 /dev/sdc1

^ would this be correct? Or am i missing something here?

clvn · 2011-02-24 14:56:34

One last bump. If i dont receive any feedback about the command above this post, i'll run it. Maybe it will solve all my problmems, maybe it will corrupt a lot of data

pyther · 2011-02-24 15:46:41

I could be wrong but errors such as these

Buffer I/O error on device sde1, logical block 983126742

usually indicates a bad drive.

clvn · 2011-02-24 16:03:39

Yes, the drive is bad, but the data is intact. That's why i dd'ed the data to sda1, and want to force the raid up with sda as one of the five out of six drives. Thanks for your imput though!

colyte · 2011-03-17 22:33:58

Re-partition sda so that it also starts at 63. dd the data over again.
Mark the sde1 as faulty and add sda1. Run a check on the array afterwards.

Arch Linux

#1 2011-02-20 20:59:45

mdadm - dd'ed a partition. "no superblock detected"

#2 2011-02-22 01:13:13

Re: mdadm - dd'ed a partition. "no superblock detected"

#3 2011-02-24 14:56:34

Re: mdadm - dd'ed a partition. "no superblock detected"

#4 2011-02-24 15:46:41

Re: mdadm - dd'ed a partition. "no superblock detected"

#5 2011-02-24 16:03:39

Re: mdadm - dd'ed a partition. "no superblock detected"

#6 2011-03-17 22:33:58

Re: mdadm - dd'ed a partition. "no superblock detected"

Board footer