dude67 Posted November 20, 2008 Report Share Posted November 20, 2008 Again I turn to you with a problem. I have a dual boot with Mandriva 2009.0 KDE4 and WinXP home. I also have a RAID1 array of two 750 GB disks - only the winXP is left out as I've got software raid built with Mandriva. Then the issue at hand: I turned Mandy on today after work and downloaded some updates (at least urpmi and then gurpmi and bunch of other stuff). I don't remember what they were. Everything was just fine at that time. Then I booted to windows as my son wanted to play some games and no problem there either. But when I rebooted back to Mandriva I couldn't get Mandriva to load anymore. Finally I booted to failsafe and was brought to shell. There was a problem with filesystem, at least this much I could catch from the screen: Invalidating stale software suspend images... done. Checking filesystems fsck.ext3: No such file or directory while trying to open /dev/md1 /dev/md1: 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> fsck.ext3: No such file or directory while trying to open /dev/md2 /dev/md2: 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> [FAILED] *** An error occured during the file system check. *** Dropping you to a shell: the system will reboot *** when you leave the shell. Give root password for maintenance (or type Contro-D to continue): As I said, I have RAID 1 array and the only thing I could make of this is that there seems to be a problem with md1 and md2. This leads me to suspect something's not right with the raid system. I saw an mdadm update note in the advisories section paul posts: (https://mandrivausers.org/index.php?showtopic=74454&hl=), but I cannot remember if that was one of the updates I downloaded. I did update everything that was suggested - just don't remember if I saw that one on my list. Anyway, are there any advice as to what to do next? [moved from Installing Mandriva by spinynorman] Quote Link to comment Share on other sites More sharing options...
ianw1974 Posted November 20, 2008 Report Share Posted November 20, 2008 First, do you remember which partitions were a part of /dev/md1 and /dev/md2? I cannot stress how important this is, else we could screw it all up. If you have access to /etc/fstab, check this. If so, then what you need to do is something like this: 1. Boot a LiveCD. 2. When booted, we need to generate an /etc/mdadm.conf file for your disks. This is easy enough and will be something like: DEVICE /dev/sda* DEVICE /dev/sdb* ARRAY /dev/md1 devices=/dev/sda2,/dev/sdb2 ARRAY /dev/md2 devices=/dev/sda3,/dev/sdb3 MAILADDR root@localhost that's assuming that /dev/sda2/sdb2 was part of md1, and sda3/sdb3 part of md2. 3. Once you've done this, check that the /dev/md* nodes exist using normal directory command: ls /dev/md* If not, you have to do it like this: mknod /dev/md1 b 9 1 mknod /dev/md2 b 9 2 4. Assemble the arrays: madam --assemble --scan 5. Check /proc/mdstat that the arrays are active. 6. Check for errors: fsck /dev/md1 fsck /dev/md2 7. Mount the partitions to check all is OK. Then reboot into Mandy again. report back with some results or problems. Quote Link to comment Share on other sites More sharing options...
dude67 Posted November 21, 2008 Author Report Share Posted November 21, 2008 Thanks Ian for helping out! OK, I'm starting in shell mode as I can get there when booting failsafe. Now I have root privileges. Then I'm assuming that you meant that I really write '/dev/sda*' and not '/dev/sda6' in the first line. I do remember my partitions: md0 = / md1 = /home md2 = /storage md0 was (and is) sda6 and sdc6 md1 was sda8 and sdc8 md2 was sda9 and sdc9 But now I can only see md0 in my /proc/mdstat: md0 : active raid1 sda6[0] sdc6[1] 20474688 blocks [2/2] [UU] unused devices: <none> I added these lines to /etc/mdadm.conf DEVICE /dev/sda* DEVICE /dev/sdc* ARRAY /dev/md1 devices=/dev/sda8,/dev/sdc8 ARRAY /dev/md2 devices=/dev/sda9,/dev/sdc9 MAILADDR root@localhost I couldn't see the md1 of md2 with 'ls /dev/md*' so I rebooted. Quote Link to comment Share on other sites More sharing options...
dude67 Posted November 21, 2008 Author Report Share Posted November 21, 2008 (edited) When rebooted I got Mandriva back! I don't know what went wrong with the system updates (or something else), but now this made it work just fine! Perhaps when updating the mdadm package the '/etc/mdadm.conf' file was overwritten and replaced by a std new file! Thanks a million Ian - you're the man! Let me buy you (again) a whole bunch of virtual beers! This is my current array: $ cat /proc/mdstat Personalities : [raid1] [raid6] [raid5] [raid4] md2 : active raid1 sdc9[1] sda9[0] 506272256 blocks [2/2] [UU] md1 : active raid1 sda8[0] sdc8[1] 102398208 blocks [2/2] [UU] md0 : active raid1 sda6[0] sdc6[1] 20474688 blocks [2/2] [UU] unused devices: <none> Edited November 21, 2008 by dude67 Quote Link to comment Share on other sites More sharing options...
ianw1974 Posted November 22, 2008 Report Share Posted November 22, 2008 Looking good Normally, if you didn't see them, the mknod to create them will then allow you to use the assemble and scan command to activate the array. Then they can be mounted in the usual way. But glad it's back up and running. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.