need help with RAID1 EFI GPT disks

[Michael Butash]

Looks like under knoppix it's not finding the disks you seek in at least 
the right order, or all the partitions you're expecting:

On 02/04/2014 11:00 AM, George Toft wrote:
> Disk /dev/sda: 640.1 GB, 640135028736 bytes
> /dev/sda1   *           1          64      512000   83  Linux
> /dev/sda2              64       77826   624618496   8e  Linux LVM
> Disk /dev/sdb: 3000.6 GB, 3000591900160 bytes
> /dev/sdb1               1      267350  2147483647+  ee  GPT
> Disk /dev/sdc: 3000.6 GB, 3000591900160 bytes
> /dev/sdc1               1      267350  2147483647+  ee  GPT
> [root at localhost ~]# 

It's not finding the partitions for the data, or the disk as you're 
trying to use parted to print or manipulate the data with mdraid.

[root at localhost ~]# mdadm --assemble --scan
mdadm: No arrays found in config file or automatically
#
## it's not finding partition type fd for linux auto raid to start 
automagically

[root at localhost ~]# mdadm --create /dev/md0 -n2 -l1 /dev/sdb3 missing
mdadm: cannot open /dev/sdb3: No such file or directory
[root at localhost ~]# mdadm --create /dev/md0 -n2 -l1 /dev/sdc3 missing
mdadm: cannot open /dev/sdc3: No such file or directory
#
## none of these are present per above

[root at localhost ~]# mdadm --create /dev/md0 -n2 -l1 /dev/sdb1 missing
mdadm: cannot open /dev/sdb1: No such file or directory
#
## this looks like you made the disk disappear from above, confirm with 
the lsblk what is there at the moment...

[root at localhost ~]# mdadm --create /dev/md0 -n2 -l1 /dev/sdc1 missing
mdadm: /dev/sdc1 appears to contain an ext2fs file system
     size=204800K  mtime=Sun Dec 29 10:32:43 2013
mdadm: Note: this array has metadata at the start and
     may not be suitable as a boot device.  If you plan to
     store '/boot' on this device please ensure that
     your boot-loader understands md/v1.x metadata, or use
     --metadata=0.90
Continue creating array? y
mdadm: Defaulting to version 1.2 metadata
mdadm: array /dev/md0 started.
#
## hmm, you just initialized that with a superblock for mdraid, probably 
nuking your ext2 boot partition here...

[root at localhost ~]# mount /dev/md0 /mnt/raid
mount: you must specify the filesystem type
[root at localhost ~]# mount -t ext4 /dev/md0 /mnt/raid
mount: wrong fs type, bad option, bad superblock on /dev/md0,
        missing codepage or helper program, or other error
        In some cases useful info is found in syslog - try
        dmesg | tail  or so
#
## you haven't created a file system on this mdraid volume at sdc1, use 
mkfs.ext4 for this on /dev/md0 first, as it needs a fs of some flavour...

It seems like I may be able to mount /dev/sdc3 if I can correct the size.
Thoughts?
#
## unfortunately it looks like you might have nuked the data, or it was 
nuked with the reboot originally on that disk as the partition is still 
simply not showing up on it, now neither side of the disk seems 
present.  If it doesn't see partition 3 in any kind of block/partition 
descriptor, it simply isn't there...

Really, you need to be super-careful with these as depending on the 
kernel or setup of udev, it tends to probe these out of order between 
boots.  You always want to either validate it by it's UUID (using "ls 
-la /dev/disk/by-uuid/), or verifying with hdparm -i as well the devices 
for which disk is what.  The metadata version mdraid is using is 
relevant as old versions are not forward compatible, and depending on 
dist they may not be able to interpret what you have/had.  Fact it's 
creating them at 1.2 metadata descriptors for the superblock, it 
*should* be able to read the old .90 or .92 blocks, but I'm almost 
thinking the disk is either damaged, or somewhere along these adventures 
you inadvertently nuked the partitions.

If you're not seeing the raided partition anymore, it's gone, but I 
think your bigger issue is the old sdb disk is toast beyond just the 
partitions.

-mb