JFS on Raid50 wont mount in FC4
 

Had a raid 50 set up last week on a new server and it seemed to work. I set it up from a pair of 3 disk raid 5 arrays and put a JFS filesystem on it. So far all good. I even put data on it (that I'd like to keep!).

Today when I booted up the machine (possibly the first reboot since I put data on the raid) I found that i could not mount /dev/md2. I get:

mount: wrong fs type, bad option, bad superblock on /dev/md2,
missing codepage or other error
In some cases useful info is found in syslog - try
dmesg | tail or so

cat /proc/mdstat shows it running:
Personalities : [raid0] [raid5]
md2 : active raid0 md0[0] md1[1]
639804288 blocks 64k chunks

md1 : active raid5 sde2[0] sdg2[2] sdf2[1]
319902208 blocks level 5, 128k chunk, algorithm 2 [3/3] [UUU]

md0 : active raid5 sdb2[0] sdd2[2] sdc2[1]
319902208 blocks level 5, 128k chunk, algorithm 2 [3/3] [UUU]

unused devices: <none>

and I've set up /etc/fstab with an entry that worked last time I booted the machine.

However mdadm --examine /dev/md2 shows:
mdadm: No super block found on /dev/md2 (Expected magic a92b4efc, got 00001fed)

Is this bad? I never checked before so I don't know. The other thing that concerned me was that mdadm --examine on /dev/md1 and /dev/md0 (the raid 5s) shows them to be raid level "raid0"???

Any suggestions on how to proceed from here without blowing away the JFS system that sits on this raid?

i dont think you can do what your thinking, so i'm not sure about data recovery

Clarification:

I did actually reboot once (i remember not being silly enough not to test this) while some data was on it and that was fine. When I boot up, md0 and md1 are autodetected and start, but md2 does not - but i can simply do a mdadm --assemble /dev/md2 /dev/md[01] and then mount /dev/md2 and it was working. But now it starts and will not mount.

Does anyone have any ideas? Why would a JFS filesystem not mount? I can only assume that something weird is going on in the underlying raid which has corrupted the filesystem? No data has been written to the disks between it working and not working, so in theory the FS is still intact.

Resolution
 

Wow - ok, found the answer everybody. Did a jfs_fsck /dev/md2, it told me the filesystem was clean, then it mounted straight away! Dunno why - but it worked.