Lacie 5Big2 Failed After Firmware Update

[unlucky1]

Okay, I took out the HDs and saw that these I didn't label! What?! I always have in the past. Anyway, I only took out the HDs once to put them in the new Lacie and knowing I had to have the order right, I did it one by one. Now putting it back to the original Lacie I am very sure I put them back right as I did it carefully 1 by 1 so if I messed it up then the mess-up would not be a random order but more like 5 4 3 2 1 and again, not likely.

Okay, installed the new mdadm:

Code: Select all

root@(none):/ # mdadm --examine /dev/sd[abcde]2
/dev/sda2:
          Magic : a92b4efc
        Version : 1.0
    Feature Map : 0x0
     Array UUID : 4f4a8ea3:4c99b769:e18e6826:872bc27f
           Name : (none):4  (local to host (none))
  Creation Time : Thu Jan  1 00:00:39 1970
     Raid Level : raid5
   Raid Devices : 5

 Avail Dev Size : 1949474544 (929.58 GiB 998.13 GB)
     Array Size : 7797895168 (3718.33 GiB 3992.52 GB)
  Used Dev Size : 1949473792 (929.58 GiB 998.13 GB)
   Super Offset : 1949474800 sectors
          State : clean
    Device UUID : 4cd37662:2ba33322:fa552e2d:896e2c95

    Update Time : Tue Jun 18 10:35:58 2013
       Checksum : f4a5a296 - correct
         Events : 51808

         Layout : left-symmetric
     Chunk Size : 512K

   Device Role : Active device 2
   Array State : AAAAA ('A' == active, '.' == missing)
mdadm: No md superblock detected on /dev/sdb2.
mdadm: No md superblock detected on /dev/sdc2.
/dev/sde2:
          Magic : a92b4efc
        Version : 1.0
    Feature Map : 0x0
     Array UUID : 4f4a8ea3:4c99b769:e18e6826:872bc27f
           Name : (none):4  (local to host (none))
  Creation Time : Thu Jan  1 00:00:39 1970
     Raid Level : raid5
   Raid Devices : 5

 Avail Dev Size : 1949474544 (929.58 GiB 998.13 GB)
     Array Size : 7797895168 (3718.33 GiB 3992.52 GB)
  Used Dev Size : 1949473792 (929.58 GiB 998.13 GB)
   Super Offset : 1949474800 sectors
          State : clean
    Device UUID : b7d90b31:1b4b0e9a:c3f998a6:fbe87a36

    Update Time : Tue Jun 18 10:35:58 2013
       Checksum : 55ce6f2e - correct
         Events : 51808

         Layout : left-symmetric
     Chunk Size : 512K

   Device Role : Active device 3
   Array State : AAAAA ('A' == active, '.' == missing)


[unlucky1]

And now one by one for my own clarity

Code: Select all

root@(none):/ # mdadm --examine /dev/sda2
/dev/sda2:
          Magic : a92b4efc
        Version : 1.0
    Feature Map : 0x0
     Array UUID : 4f4a8ea3:4c99b769:e18e6826:872bc27f
           Name : (none):4  (local to host (none))
  Creation Time : Thu Jan  1 00:00:39 1970
     Raid Level : raid5
   Raid Devices : 5

 Avail Dev Size : 1949474544 (929.58 GiB 998.13 GB)
     Array Size : 7797895168 (3718.33 GiB 3992.52 GB)
  Used Dev Size : 1949473792 (929.58 GiB 998.13 GB)
   Super Offset : 1949474800 sectors
          State : clean
    Device UUID : 4cd37662:2ba33322:fa552e2d:896e2c95

    Update Time : Tue Jun 18 10:35:58 2013
       Checksum : f4a5a296 - correct
         Events : 51808

         Layout : left-symmetric
     Chunk Size : 512K

   Device Role : Active device 2
   Array State : AAAAA ('A' == active, '.' == missing)
root@(none):/ # mdadm --examine /dev/sdb2
mdadm: No md superblock detected on /dev/sdb2.
root@(none):/ # mdadm --examine /dev/sdc2
mdadm: No md superblock detected on /dev/sdc2.
root@(none):/ # mdadm --examine /dev/sdd2
mdadm: cannot open /dev/sdd2: No such file or directory
root@(none):/ # mdadm --examine /dev/sde2
/dev/sde2:
          Magic : a92b4efc
        Version : 1.0
    Feature Map : 0x0
     Array UUID : 4f4a8ea3:4c99b769:e18e6826:872bc27f
           Name : (none):4  (local to host (none))
  Creation Time : Thu Jan  1 00:00:39 1970
     Raid Level : raid5
   Raid Devices : 5

 Avail Dev Size : 1949474544 (929.58 GiB 998.13 GB)
     Array Size : 7797895168 (3718.33 GiB 3992.52 GB)
  Used Dev Size : 1949473792 (929.58 GiB 998.13 GB)
   Super Offset : 1949474800 sectors
          State : clean
    Device UUID : b7d90b31:1b4b0e9a:c3f998a6:fbe87a36

    Update Time : Tue Jun 18 10:35:58 2013
       Checksum : 55ce6f2e - correct
         Events : 51808

         Layout : left-symmetric
     Chunk Size : 512K

   Device Role : Active device 3
   Array State : AAAAA ('A' == active, '.' == missing)


[Mijzelf]

I'm missing sdd2. I suppose you have to re-create the device nodes /dev/sdd and /dev/sdd2.

Out of that it seems mdadm wasn't the problem.

[unlucky1]

Code: Select all

root@(none):/ # mdadm --examine /dev/sdd2
/dev/sdd2:
          Magic : a92b4efc
        Version : 1.0
    Feature Map : 0x0
     Array UUID : 4f4a8ea3:4c99b769:e18e6826:872bc27f
           Name : (none):4  (local to host (none))
  Creation Time : Thu Jan  1 00:00:39 1970
     Raid Level : raid5
   Raid Devices : 5

 Avail Dev Size : 1949474544 (929.58 GiB 998.13 GB)
     Array Size : 7797895168 (3718.33 GiB 3992.52 GB)
  Used Dev Size : 1949473792 (929.58 GiB 998.13 GB)
   Super Offset : 1949474800 sectors
          State : clean
    Device UUID : 9134e88d:390f3c7e:ae8566b5:7a779902

    Update Time : Tue Jun 18 10:35:58 2013
       Checksum : 71c4a891 - correct
         Events : 51808

         Layout : left-symmetric
     Chunk Size : 512K

   Device Role : Active device 4
   Array State : AAAAA ('A' == active, '.' == missing)


[unlucky1]

So far we have 2 disks that say RAID5 and 2 that say nothing and 1 that says Clean, right? The goal is to have 5 that say RAID5 but we can settle for 4- is my understanding right here?

[fvdw]

its lunch time

from what you posted it seems you have this
sda2 raid 5, clean, active device 2
sdb2, no superblock
sdc2, no superblock
sdd2, raid 5, clean active device 4
sde2, raid 5, clean, active divice 3

[unlucky1]

Yes, I gotta run some errands. Be back soon!

[fvdw]

what about a test to swap disk to see if the problem moves with the disk, this to rule out a hardware problem

Think as long as we don't trbuild the array it can not harm, correct Mijzelf ?

If so i would swap the disk in slot 3 with that in slot 5 and see if now sde hasno superblock.

If that is the case then the problem is with the disks.

If the postion of the raid header is know we can copy it fromthat sector as described earlier and have alook at it with a hex editor

[Mijzelf]

Yes, you are right, besides that we have 3 clean devices. (is an answer on post 73)

From an old post of you:

Code: Select all

 8 0 976762584 sdc
 8 16 976762584 sdb
 8 32 976762584 sda
 8 48 976762584 sde
 8 64 976762584 sdd

Here the names are the names as used by the kernel, which indicates the position in the nas. I *think* that sda is left, when seen from the back. But maybe it's right. fvdw will know that, I suppose.

The minor (second number, the first one is the major) shows the names we are using. 0=a, 16=b, ...
Further we have the device role as told by mdadm,
sda=2, sdd=4, sde=3.

When I combine that, I get:

Code: Select all

Physical position username role
1                 sdc      ?
2                 sdb      ?
3                 sda      2
4                 sde      3
5                 sdd      4

So we are missing role 0 and 1, and if you didn't screw up the disk order, that is sdc and sdb.

Well, let's first try

Code: Select all

mdadm --assemble /dev/md0 --uuid=4f4a8ea3:4c99b769:e18e6826:872bc27f

If that works, do a

Code: Select all

cat /proc/mdstat

If that fails I think we'll have to re-create the array, in this order.

/Edit: If that fails it might be a good idea to first bring the 'userspace names' in the same order as the roles (and the kernel names):

Code: Select all

rm /dev/sd*

mknod -m 666 /dev/sdc b 8 0
mknod -m 666 /dev/sdc2 b 8 2

mknod -m 666 /dev/sdb b 8 16
mknod -m 666 /dev/sdb2 b 8 18

mknod -m 666 /dev/sda b 8 32
mknod -m 666 /dev/sda2 b 8 34

mknod -m 666 /dev/sde b 8 48
mknod -m 666 /dev/sde2 b 8 50

mknod -m 666 /dev/sdd b 8 64
mknod -m 666 /dev/sdd2 b 8 66

And retry the assembly on uuid.

[Mijzelf]

Think as long as we don't trbuild the array it can not harm,, correct Mijzelf ?

Correct.