Re: [Solved] Trouble creating/assembling/mounting Ver 1.0 raid1 disk in another box to recover data

[Wols Lists <antlists@youngman.org.uk>]

On 25/10/17 22:05, David C. Rankin wrote:
> Neil, All,

Hi,

I'll just add a few hints for troubleshooting next time :-)
> 
> (sometimes it's the painfully obvious that just slips right by... I was going
> to ask for help, but discovered my problem while going through and composing
> the information for this e-mail -- kind of funny, so I thought I would pass it
> along... and maybe help someone else in the future)
> 
>   One of my long time boxes finally reached end of life due to a capacitor
> issue a couple of days ago. I had "almost" everything I needed off the raid1
> array before it finally died, but I missed getting the faxes out of the
> hylafax/avantfax folder and I need to recover that data.
> 
>   I have the drive attached to another box via a ATA/usb cable and the drive
> is recognized, etc., but when I try to create an array with the drive (and
> missing) to be able to mount it somewhere, it always fails with "Device Busy".
> 
> (I don't know what "Busy" means here, but I suspect that when I try and
> create/assemble it, it used to be a md0, md1, or md2 on the old box and this
> current box has a md0-4 already. -- I don't know if that makes any sense, but
> that is all I could think of off-hand)

Typically, "Busy" means it's already assembled into an array, as you
found out ...
> 
> <addendum -- I was wrong, keep reading :)>
> 
>   The drive is an older ATA drive and I don't have a box with ATA anymore, but
> I do have a ATA/usb adapter from the time when laptop drives were ATA.
> Unfortunately the power supply for the laptop drive won't spin the 3.5 inch
> disk, so I have the power-supply from another computer sitting on the table,
> plugged in, with the green and back wire from the 24-pin connector jumped to
> start the power supply. The drive is sde in the current partition list:
> 
> # cat /proc/partitions
> major minor  #blocks  name
> 
>   11        0    1048575 sr0
>    8        0  976762584 sda
>    8        1          1 sda1
>    8        5     512000 sda5
>    8        6   52428800 sda6
>    8        7  921161728 sda7
>    8        8    2117632 sda8
>    8       16  976762584 sdb
>    8       17          1 sdb1
>    8       21     512000 sdb5
>    8       22   52428800 sdb6
>    8       23  921161728 sdb7
>    8       24    2117632 sdb8
>    8       32 2930266584 sdc
>    8       48 2930266584 sdd
>    9        4 2930135488 md4
>    9        3    2115584 md3
>    9        1   52396032 md1
>    9        2  921030656 md2
>    9        0     511680 md0
>    8       64  245117376 sde
>    8       65     104391 sde1
>    8       66          1 sde2
>    8       69   20972826 sde5
>    8       70    2104483 sde6
>    8       71  221929911 sde7
> 
> 
>   The drive data is all in tact from the motherboard's time of death on Oct
> 20, e.g.
> 
> $ mdadm -E /dev/sde5
> /dev/sde5:
>           Magic : a92b4efc
>         Version : 1.0
>     Feature Map : 0x1
>      Array UUID : e45cfbeb:77c2b93b:43d3d214:390d0f25
>            Name : 1
>   Creation Time : Thu Aug 21 01:43:22 2008
>      Raid Level : raid1
>    Raid Devices : 2
> 
>  Avail Dev Size : 41945504 (20.00 GiB 21.48 GB)
>      Array Size : 20972752 (20.00 GiB 21.48 GB)
>    Super Offset : 41945632 sectors
>    Unused Space : before=0 sectors, after=47 sectors
>           State : clean
>     Device UUID : e0c1c580:db4d853e:6fac1c8f:fb5399d7
> 
> Internal Bitmap : -81 sectors from superblock
>     Update Time : Fri Oct 20 15:55:58 2017
>        Checksum : db3d3417 - correct
>          Events : 19154344
> 
> 
>    Device Role : Active device 0
>    Array State : AA ('A' == active, '.' == missing, 'R' == replacing)
> 
>   When I try and create a new array (say md20) out of this drive, it just says
> "Device Busy", e.g.
> 
> # mdadm --verbose --create /dev/md20 --level=1 --raid-devices=2 /dev/sde5 missing
> mdadm: super1.x cannot open /dev/sde5: Device or resource busy
> mdadm: ddf: Cannot use /dev/sde5: Device or resource busy
> mdadm: Cannot use /dev/sde5: It is busy
> mdadm: cannot open /dev/sde5: Device or resource busy
> 
>   There is no md20 to conflict with:
> 
> # l /dev/md20
> ls: cannot access '/dev/md20': No such file or directory
> 
>   The only md devices on this current system are:
> 
> l /dev/md*
> brw-rw---- 1 root disk 9,   0 Oct 23 01:25 /dev/md0
> brw-rw---- 1 root disk 9,   1 Oct 23 01:25 /dev/md1
> brw-rw---- 1 root disk 9, 125 Oct 25 14:26 /dev/md125
> brw-rw---- 1 root disk 9, 126 Oct 25 14:26 /dev/md126
> brw-rw---- 1 root disk 9, 127 Oct 25 14:26 /dev/md127
> brw-rw---- 1 root disk 9,   2 Oct 23 01:25 /dev/md2
> brw-rw---- 1 root disk 9,   3 Oct 23 01:25 /dev/md3
> brw-rw---- 1 root disk 9,   4 Oct 23 01:25 /dev/md4
> 
> 
> ...SMACKS SELF FOR BEING SO DUMB... (will place the pointy hat on head, and
> turn and face the corner when this message is done)
> 
>   Duh! Where did md125-127 come from? Those are not normally there. Crap! When
> I plugged the usb in, since mdadm was already running on the box, it saw the
> drive at the end of my usb cable as an array and had already assembled -- but
> not started the array --> which explains the "Device Busy" error.

If arrays are not named, then mdadm by default now counts down from 127
... I'm slightly surprised --examine didn't tell you what array it was
part of.
> 
> # mdadm -D /dev/md126
> /dev/md126:
>         Version : 1.0
>      Raid Level : raid0
>   Total Devices : 1
>     Persistence : Superblock is persistent
> 
>           State : inactive
> 
>            Name : 1
>            UUID : e45cfbeb:77c2b93b:43d3d214:390d0f25
>          Events : 19154344
> 
>     Number   Major   Minor   RaidDevice
> 
>        -       8       69        -        /dev/sde5
> 
> 
>   I can't believe I wasted 20 minutes dorking with the getting the darn thing
> assembled, but didn't snap to the fact it might already be assembled. Nothing
> in my wildest expectations said plugging a usb cable from a bare drive laying
> on the table powered by a jump-started PS would automatically be recognized as
> an array and assembled -- but it did. Damn good job guys!
> 
ALWAYS try to stop an array before trying the next thing to get it
started - a failed (or in this case succeeded :-) assembly always messes
up further attempts at diagnosis :-)

> # mdadm --verbose --readonly --run /dev/md126
> mdadm: started array /dev/md/1_0
> 
>   Hah! good!
> 
> # mdadm -D /dev/md126
> /dev/md126:
>         Version : 1.0
>   Creation Time : Thu Aug 21 01:43:22 2008
>      Raid Level : raid1
>      Array Size : 20972752 (20.00 GiB 21.48 GB)
>   Used Dev Size : 20972752 (20.00 GiB 21.48 GB)
>    Raid Devices : 2
>   Total Devices : 1
>     Persistence : Superblock is persistent
> 
>   Intent Bitmap : Internal
> 
>     Update Time : Fri Oct 20 15:55:58 2017
>           State : clean, degraded
>  Active Devices : 1
> Working Devices : 1
>  Failed Devices : 0
>   Spare Devices : 0
> 
>            Name : 1
>            UUID : e45cfbeb:77c2b93b:43d3d214:390d0f25
>          Events : 19154344
> 
>     Number   Major   Minor   RaidDevice State
>        0       8       69        0      active sync   /dev/sde5
>        -       0        0        1      removed
> 
>   Hah! Double Good!
> 
> # mount /dev/md126 /mnt/af/
> # md /home/data/nemesis/srv/
> # cd /mnt/af/srv/
> # cp -a avantfax/faxes /home/data/nemesis/srv/
> # cp -a avantfax/tmp /home/data/nemesis/srv/
> 
>   Bingo! Done!
> 
>   Thanks a lot for all the help -- really. If I hadn't been toiling through
> the process of collecting all the information to prepare a well thought out
> post for the list, who knows how long it would have taken me to stumble across
> the fact that the arrays had already been assembled.
> 
>   So hopefully this brings a chuckle to all, and maybe some help to someone
> similarly situated in the future (dunce hat and all)
> 
Oh the joys of having the computer being too clever by half :-)

Cheers,
Wol