I've spent a number of hours over the last week attempting to set up my new desktop with software RAID1. It's a Debian Sid install from a "fresh" daily image of the installer.

I managed to get the RAID1 set up using mdadm and things were working well for a couple of days. Yesterday, I installed a number of packages (I don't know if this had anything to do with it or not) and rebooted and upon reboot I was greeted with fsck failures on ALL of my RAID devices. The superblocks were all corrupted or "gone". It suggested that I run fs2ck (all of the block devices are ext3) to load an alternate superblock, but I received the same error again... corrupted or no superblock...

At that point I was left with no choice but to reinstall.

Does anyone have any ideas as to why this happened? What could I have installed to "break" things? Does anyone know what I can do to prevent this?

I reinstalled but simply partitioned my second disk into a single partition and mounted it as /backup and I will either use a backup solution using it for now...
Right now, I don't trust software RAID1 to be stable enough for me to use everyday!!!!

Thanks in advance for your thoughts!

Correction! Here was the exact error that I received:

The superblock could not be read or does not describe a correct ext2
filesystem. If the device is valid and it really contains an ext2
filesystem (and not swap or ufs or something else), then the superblock
is corrupt, and you might try running e2fsck with an alternate superblock:
e2fsck -b 8193 <device>

be sure all of your partitions you use for raid are id'd to "fd" (=raid autodetect) so it automatically get mounted at boot. check with ex. fdisk /dev/sdx, chande with "t" .

sl mritch.

All of the Raid1 partitions were set to fd (linux raid auto-detect) The raid had been working fine and had been rebooted a number of times over several days and then wham, a reboot and fsck failure! ALL of the md superblocks were corrupted and unrecoverable!

Thoughts or ideas??? Right now I have given up on being able to get a RAID1 system and I will be using partimage to backup my partitions to the now unused second disk... But I would set up the raid again if I could be sure of stability and problem free operation (or at least recoverable!).

Thanks for your thoughts!

well. i'm using raid0 & raid1 software with an ext2 and reiserfs above.
it's ver. 0.90.20010914-15 (bit older sarge) with kernel 2.4.26(+scsi-idle/preempt/lowlat), several scsi + two ide disks.

no probs shown up so far.
maybe try it on two small partitions and stress it. for me that rather sounds like a detection problem. something other broken (initscripts?).

sl mritch.

That's a good idea! I could set up a couple of partitions before going with the whole thing to see how things work out!

When I had the Raid1 "fail" I had been installing quite a few packages. I don't "think" that had anything to do with it, but...

Thanks for the idea!

We had that error and it turned out to be a kernel which didn't support RAID. Could this be a possibility?

Cheers,

Enkidu

You didn't mention if you were installing those packages to your raid or not.
Installing packages shouldn't have affected your raid at all.
Have you checked your fstab, maybe it got corrupted