Re: IMSM Raid 5 always read only and gone after reboot

On 08/16/11 13:19, Iwan Zarembo wrote:

Hello Everyone,
I am not that new to Linux, but I am quite far away from being an
expert :) I am using linux for a few years now and everything worked
just fine, but not this one with IMSM Raid. I googled for some weeks
and asked everyone I know about the problem but without any luck. The
last possibility to find a solution is this mailing list. I realy hope
someone can help me.

I've just gone through this myself.

(snip)

I started the raid by entering the command:
    # mdadm -I -e imsm /dev/md127
    mdadm: Started /dev/md/raid with 4 devices

Now mdstat has the following output:
    # cat /proc/mdstat
Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5]
[raid4] [raid10]
md126 : active (read-only) raid5 sdd[3] sdc[2] sdb[1] sde[0]
      2930280448 blocks super external:/md127/0 level 5, 128k chunk,
algorithm 0 [4/4] [UUUU]
      	resync=PENDING

md127 : inactive sde[3](S) sdb[2](S) sdc[1](S) sdd[0](S)
      9028 blocks super external:imsm

unused devices: <none>

What you are seeing here is the imsm container (/dev/md127), which you
generally don't use unless you are trying to reconfigure arrays.

The other device (/dev/md126) is the actual raid5 array as defined in
the imsm BIOS. This is what you use in disk operations. Examples:

$ parted /dev/md126

Then create partitions on the device - you'll have to use something
compatible with gpt tables. When you do this, you'll have new devices
available to you, such as /dev/md126p1 (first partition), /dev/md126p2
(second partition), and so on.

However, if all you're doing is creating one big partition, you don't
necessarily need a partition table, you can create a filesystem right on
the array itself (/dev/md126) as I did on my server.

I learned that md126 is so long read only until it was used the first
time. So I tried to create a partition with the documentation from the
wiki, but not with ext3. I used ext4 for this.

 #  mkfs.ext4 -v -m .1 -b 4096 -E stride=32,stripe-width=64 /dev/md/raid
The result was:
mke2fs 1.41.14 (22-Dec-2010)
fs_types für mke2fs.conf Lösung: 'ext4'
/dev/md/raid: The operation is not allowed then creating Superblocks.
Original message in German: Die Operation ist nicht erlaubt beim
Erstellen des Superblocks

This is the first problem. I am not able to do anything on the raid drive.

Try using:

`mkfs.ext4 -v -m .1 -b 4096 -E stride=32,stripe-width=64 /dev/md126`

This, of course, assumes you aren't using a partition table. Substitute
the correct partition device should you create one.

The second problem is that the raid is gone after the reboot!

I'm not familiar with ubuntu, but you likely need to add a service to
the startup scripts in order to start the raid array and have it usable,
assuming that this is not the root device and is just being used for
storage. It may need kernel arguments to tell mdadm to find and assemble
arrays. Hopefully someone with Ubuntu experience can answer this.

I know on my distribution (not Ubuntu!) I have to add a service to the
boot runlevel in order to assemble arrays that are not the root filesystem.

Dan
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html