mdadm Device or resource busy using drives from old harware raid

[oehqparts]

Hello all ,
I am trying to set up a mdadm raid in a new PC that I am building for home theatre.
any help here would be greatly appreciated.

the machine boot just fine from /dev/sdc running ubuntu 9.10
However in gparted /dev/sda and dev/sdb show to be part of /devmapper/sil_ajbicfacbaej

Both dev/sda and /dev/sdb were drives that used to be part of a sil hardware raid on a previous machine. I would like to use them as a new mdadm raid on this new machine the old hardware card was really quite slow. the drives are now pluged into the MB and should bw much faster there.


fdisk -l shows this
***********************************************
~$ sudo fdisk -l

Disk /dev/sda: 1500.3 GB, 1500301910016 bytes
255 heads, 63 sectors/track, 182401 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x000cc314

Device Boot Start End Blocks Id System

Disk /dev/sdb: 1500.3 GB, 1500301910016 bytes
255 heads, 63 sectors/track, 182401 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x000cc314

Device Boot Start End Blocks Id System

Disk /dev/sdc: 250.1 GB, 250059350016 bytes
255 heads, 63 sectors/track, 30401 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x000e95c0

Device Boot Start End Blocks Id System
/dev/sdc1 * 1 29367 235890396 83 Linux
/dev/sdc2 29368 30401 8305605 5 Extended
/dev/sdc5 29368 30401 8305573+ 82 Linux swap / Solaris
************************************************
cat /proc/mdstat show the folowing

~$ cat /proc/mdstat
Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] [raid4] [raid10]
unused devices: <none>
**************************************************
Creating mdadm produces the following

~$ sudo mdadm --create --verbose /dev/md0 --level=1 --raid-devices=2 /dev/sda /dev/sdb
mdadm: Cannot open /dev/sda: Device or resource busy
mdadm: Cannot open /dev/sdb: Device or resource busy

************************************************** ******
I tried the command
sudo dmsetup remove /dev/mapper/sil_ajbicfacbaej

but it does not appear to remove it.
Any Ideas as to how to start over and use these drives in mdadm??

Thanks for your help

[Irithori]

What happens is this:
- the sil "hardware" raid (a more appropiate name would be fakeraid. I can sustain that with some arguments, if asked. But for now let´s leave it out, as it distracts from the topic at hand) has written a signature to the harddiscs sda and sdb.
This signature says: "I belong to a sil-raid"
- at boot time, the kernel gets to know sda and sdb via sata.
- the kernel also reads that signature and therefore loads and activates the sil_?? kernel module.
- that sil kernel module binds sda and sdb and creates the device /devmapper/sil_ajbicfacbaej

So the quick and dirty way to solve that raid issue is to prevent this sil_?? module to be loaded by configuring modules.conf

The quick and _very_ dirty way is to simply delete (or move) the sil_?? module in
/lib/modules/<kernel-version/
But this will break at the next kernel update.

I would advise to get rid of that signature.
Unfortunately, I do not know a commandline tool to do it from linux.
If you find one, let me know
Even dd will not help, as this signature is outside the usual data area of the harddisc.

Can you attach the two harddiscs to that sil controller again?
If yes, go to its BIOS, delete the raid and set them to "single disc" or "not raid" or whatever will produce a "normal", not-raided disc.

[oehqparts]

Thanks so much for the reply and insight.

Problem solved...

I solved the problem by booting from a freedos CD and then running fdisk under freedos. (freedos is available via download at "freedos.org"

The Freedos version of fdisk has a option to delete a unknown partition. After I deleted the partitions I was able to run

sudo dmsetup remove /dev/mapper/sil_ajbicfacbaej
which totaly remover the sil fakerade

Then when I ran mdadm and all was well...

Thanks again for all the help just needed a better understanding and a pion in the right direction.

[oehqparts]

Problem Back....I thought I solved this problem by removing the partitions with freedos.

it worked for awhile but after my first reboot my mdadm raid dissapeared and the
/dev/mapper/sil_ajbicfacbaej shows up again in my listing

************************************************** ***********************
heffsvr2@heffsvr-2:~$ sudo blkid
/dev/sda: UUID="4908b0f8-ea1a-58d0-2e04-d0d500234044" TYPE="linux_raid_member"
/dev/sdb: UUID="4908b0f8-ea1a-58d0-2e04-d0d500234044" TYPE="linux_raid_member"
/dev/sdc1: UUID="d8b53af3-55c9-4a1a-acb9-f3cb39096434" TYPE="ext3"
/dev/sdc5: UUID="7ec97044-9864-46b0-b925-7fa40aa4a0a9" TYPE="swap"
/dev/mapper/sil_ajbicfacbaej: UUID="c63b7862-d538-494f-94bf-80ab085f5293" TYPE="ext3"
/dev/sdd1: LABEL="TOSHIBA EXT" UUID="162E-1B0A" TYPE="vfat"
************************************************** ***************************
does anyone know how to permanently get rid of the
/dev/mapper/sil_ajbicfacbaej

here is the listing from fdisk -l
************************************************** **************
heffsvr2@heffsvr-2:~$ sudo fdisk -l

Disk /dev/sda: 1500.3 GB, 1500301910016 bytes
255 heads, 63 sectors/track, 182401 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x00000000

Disk /dev/sda doesn't contain a valid partition table

Disk /dev/sdb: 1500.3 GB, 1500301910016 bytes
255 heads, 63 sectors/track, 182401 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x00000000

Disk /dev/sdb doesn't contain a valid partition table

Disk /dev/sdc: 250.1 GB, 250059350016 bytes
255 heads, 63 sectors/track, 30401 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x000e95c0

Device Boot Start End Blocks Id System
/dev/sdc1 * 1 29367 235890396 83 Linux
/dev/sdc2 29368 30401 8305605 5 Extended
/dev/sdc5 29368 30401 8305573+ 82 Linux swap / Solaris

Disk /dev/sdd: 500.1 GB, 500107862016 bytes
255 heads, 63 sectors/track, 60801 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0xc27c4f8f

Device Boot Start End Blocks Id System
/dev/sdd1 1 60801 488383008+ c W95 FAT32 (L
************************************************** *****************************

any help is greatly appreciated!!

[Irithori]

If you already have data on the discs, make a backup first.

For the backup purpose, you can remove the sil kernel module
and reboot. This will prevent the creation of /dev/mapper/sil_ajbicfacbaej

Do the backup, then copy the sil kernel module back to its place.


To permanently get rid of /dev/mapper/sil_ajbicfacbaej
reattach the discs to the sil controller.
Go into the sil bios and delete the raid.

[oehqparts]

Thanks so much for the help!!! (you guys are the BEST!)
Both the drives are empty ( so there is no data) to loose I am running Ubuntu 9.10

I connected the drives back to the old controller and believe I have successfully deleted the /dev/mapper/sil_ajbicfacbaej.

now when I complete the mdadm set-up thing work OK until a reboot.
then I get an error stating md0 does not exist

when I run
sudo cat /proc/mdstat i get
Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] [raid4] [raid10]
md_d0 : inactive sda[0](S)
1465138496 blocks


I do not understand where the "_" came from the mdadm statement I am running is
sudo mdadm --create --verbose /dev/md0 --level=1 --raid-devices=2 /dev/sda /dev/sdb

without the underscore ...do you know why this is happening?

I can stop the raid with
sudo mdadm -S /dev/md_d0
and the reset it up with a new
sudo mdadm --create --verbose /dev/md0 --level=1 --raid-devices=2 /dev/sda /dev/sdb
and all works well again, but after reboot it is gone again

also I cannot mount the md0 after setup it is only accessible from the "1500 GB Filesystem" icon on the desktop.

Here is a screen shot of the steps I am taking after reboot
************************************************** ***************
heffsvr2@heffsvr-2:~$ sudo blkid
[sudo] password for heffsvr2:
/dev/sda: UUID="fe96b570-8155-42e0-2e04-d0d500234044" TYPE="linux_raid_member"
/dev/sdb: UUID="fe96b570-8155-42e0-2e04-d0d500234044" TYPE="linux_raid_member"
/dev/sdc1: UUID="d8b53af3-55c9-4a1a-acb9-f3cb39096434" SEC_TYPE="ext2" TYPE="ext3"
/dev/sdc5: UUID="7ec97044-9864-46b0-b925-7fa40aa4a0a9" TYPE="swap"

( NOTE dev/md0 is missing)

heffsvr2@heffsvr-2:~$ sudo cat /proc/mdstat
Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] [raid4] [raid10]
md_d0 : inactive sdb[1](S)
1465138496 blocks
unused devices: <none>

heffsvr2@heffsvr-2:~$ sudo mdadm -S /dev/md_d0
mdadm: stopped /dev/md_d0

heffsvr2@heffsvr-2:~$ sudo mdadm --create --verbose /dev/md0 --level=1 --raid-devices=2 /dev/sda /dev/sdb
mdadm: /dev/sda appears to contain an ext2fs file system
size=1465138496K mtime=Sat Jan 23 14:53:36 2010
mdadm: /dev/sda appears to be part of a raid array:
level=raid1 devices=2 ctime=Sat Jan 23 14:49:52 2010
mdadm: /dev/sdb appears to contain an ext2fs file system
size=1465138496K mtime=Sat Jan 23 14:53:36 2010
mdadm: /dev/sdb appears to be part of a raid array:
level=raid1 devices=2 ctime=Sat Jan 23 14:49:52 2010
mdadm: size set to 1465138496K
Continue creating array? y
mdadm: array /dev/md0 started.

heffsvr2@heffsvr-2:~$
heffsvr2@heffsvr-2:~$ sudo blkid
/dev/sda: UUID="a499fb4a-35f8-0e2c-2e04-d0d500234044" TYPE="linux_raid_member"
/dev/sdb: UUID="a499fb4a-35f8-0e2c-2e04-d0d500234044" TYPE="linux_raid_member"
/dev/sdc1: UUID="d8b53af3-55c9-4a1a-acb9-f3cb39096434" SEC_TYPE="ext2" TYPE="ext3"
/dev/sdc5: UUID="7ec97044-9864-46b0-b925-7fa40aa4a0a9" TYPE="swap"
/dev/md0: UUID="c499ef07-1218-4d3e-afff-12c000dd8808" TYPE="ext3"

************************************************** ************
sudo mount /dev/md0 /media/Movies
mount: /dev/md0 already mounted or /media/Movies busy
mount: according to mtab, /dev/md0 is already mounted on /media/Movies

************************************************** ******

(when I try to access Movies from the file browser I get the error the same error)

mount: /dev/md0 already mounted or /media/Movies busy
mount: according to mtab, /dev/md0 is already mounted on /media/Movies


there is a raid looking icon that shows up on the desktop called "1500 GB Filesystem"
if I click on this icon I can access the drive array.
***********************************************
heffsvr2@heffsvr-2:~$ sudo mount
/dev/sdc1 on / type ext3 (rw,relatime,errors=remount-ro)
none on /proc type proc (rw,noexec,nosuid,nodev)
none on /sys type sysfs (rw,noexec,nosuid,nodev)
none on /sys/fs/fuse/connections type fusectl (rw)
none on /sys/kernel/debug type debugfs (rw)
none on /sys/kernel/security type securityfs (rw)
udev on /dev type tmpfs (rw,mode=0755)
none on /dev/pts type devpts (rw,noexec,nosuid,gid=5,mode=0620)
none on /dev/shm type tmpfs (rw,nosuid,nodev)
none on /var/run type tmpfs (rw,nosuid,mode=0755)
none on /var/lock type tmpfs (rw,noexec,nosuid,nodev)
none on /lib/init/rw type tmpfs (rw,nosuid,mode=0755)
binfmt_misc on /proc/sys/fs/binfmt_misc type binfmt_misc (rw,noexec,nosuid,nodev)
gvfs-fuse-daemon on /home/heffsvr2/.gvfs type fuse.gvfs-fuse-daemon (rw,nosuid,nodev,user=heffsvr2)
/dev/md0 on /media/Movies type ext3 (rw,noexec,nosuid,nodev)

************************************************** ********
under the file browser both the "Movies" mount shows and the "1500 GB Filesystem" show up but only the 1500 GB Filesystem is accessible the other generates the mount error above???

and after reboot I have to do the whole thing over again???

Thanks so much for the help

[oehqparts]

[Solved]

the following link answered my issue. turns out that you must list the array in the
/etc/mdadm/mdadm.conf file

see link to solution
"ubuntuforums.org/showthread.php?t=1225334&highlight=mdadm"