Lacie 5big Network 2 - Raid Array Missing

[pfwaaa]

root@(none):/ # free
total used free shared buffers
Mem: 511672 25220 486452 0 0
-/+ buffers: 25220 486452
Swap: 0 0 0
root@(none):/ # ./xfs_repair -m 250 /dev/md4
Phase 1 - find and verify superblock...
Required memory for repair is greater that the maximum specified with the -m option. Please increase it to at least 750.
root@(none):/ #
:cry:
<boring old ****> I started on Unix V7, porting it from PDP11 to M68000. 512kBytes was a lot! </boring old ****>

Unless anyone has a really good idea, I guess I'll be bolting them onto a PC and running a distro :sob

Thanks

Peter

[Jocko]

At least there is the swap partition on sda5 (256MB) and we might use it. As it is fvdw who built the standalone kernel, it can valid this suggestion.

I don't know if a swap partition is set with the other disks if yes, we can try to create a linear raid to get a big swap partition.

Unless anyone has a really good idea, I guess I'll be bolting them onto a PC and running a distro

Yes this could be done if you can plug in all disks but you should do some additional command to be able to assemble again the raid and before trying this you MUST do a mark on all disks to be able to keep the disks order in your 5 big2.

[pfwaaa]

Progress report :)
Looking at the Lacie log files that I copied off it, Swap is built on /dev/md3 using sd[abcde]5 (as a Raid 1) - so I gave it a go after deleting everything I could:

tftp -l /sbin/mdadm -r mdadm -g 192.168.144.55
chmod 755 /sbin/mdadm
mdadm --assemble /dev/md4 /dev/sd[abcde]2
tftp -l /sbin/xfs_repair -r xfs_repair -g 192.168.144.55
chmod 755 /sbin/chmod 755 xfs_repairxfs_repair
tftp -l glibc-mini-mkfs.xfs-25feb14.tar -r glibc-mini-mkfs.xfs-25feb14.tar -g 192.168.144.55
tar -xvf glibc-mini-mkfs.xfs-25feb14.tar -C /
mdadm --assemble /dev/md3 /dev/sd[abcde]5
swapon /dev/md3
# next lines added later
rm /sbin/mdadm
rm /sbin/udevstart
rm /sbin/gdisk
rm /sbin/buttons-nwsp2
rm /bin/busybox
rm /*.tar
rm /bin/busybox
# end of added deletes
/sbinmdadm --examine /dev/sd[abcde]2 /xfs_repair -m 800 /dev/md4

I got some way:
root@(none):/ # /sbin/xfs_repair -m 750 /dev/md4
Phase 1 - find and verify superblock...
Required memory for repair is greater that the maximum specified with the -m option. Please increase it to at least 750.
root@(none):/ # /sbin/xfs_repair -m 800 /dev/md4
Phase 1 - find and verify superblock...
Phase 2 - using internal log
- zero log...
ERROR: The filesystem has valuable metadata changes in a log which needs to
be replayed. Mount the filesystem to replay the log, and unmount it before
re-running xfs_repair. If you are unable to mount the filesystem, then use
the -L option to destroy the log and attempt a repair.
Note that destroying the log may cause corruption -- please attempt a mount
of the filesystem before doing this.
root@(none):/ # mount -o ro /dev/md4 /mountpoint
mount: mounting /dev/md4 on /mountpoint failed: No such file or directory
root@(none):/ # mkdir /mountpoint
root@(none):/ # mount -o ro /dev/md4 /mountpoint
mount: mounting /dev/md4 on /mountpoint failed: Input/output error
root@(none):/ # /sbin/xfs_repair -L -m 800 /dev/md4
Phase 1 - find and verify superblock...
Phase 2 - using internal log
- zero log...
ALERT: The filesystem has valuable metadata changes in a log which is being
destroyed because the -L option was used.
- scan filesystem freespace and inode maps...
Killed
root@(none):/ #

but each time it wanted more memory than I had :cry:

So tomorrow night I need to work out how to convert Raid 1 to Raid 0. Happily mkswap is built into the standalone :applause

Thanks for the help so far
Peter

[pfwaaa]

So:
Changed Swap from Raid 1 to Raid 0. xfs_repair runs :)
Changed Swap back (I can use it with standalone so I put it back properly ;) ) and boot Lacie - Now no user interfaces and doesn't respond to the On/Off Switch :(
I currently have a *very* expensive paperweight :cry:

OK, back to the standalone and see if I can mount the Lacie disks and find its last logs...
Meanwhile getting ready to bring up a RasPi to practice with for when I finally can the Lacie software

Peter

[pfwaaa]

Ooops Forgot - Yes I can mount /dev/md4 once repair has finished with it :-D

[fvdw]

Ooops Forgot - Yes I can mount /dev/md4 once repair has finished with it :-D

at least that is something to start with ;)

[pfwaaa]

Well, I have been doing a little playing around with /dev/md4...
1) I can find all my shares and so far, all my files
2) If I try and list the files that I was attempting to delete when the Lacie died, the ls process is killed (I guess due to lack of memory)
3) tried it with the swap enlarged, just in case - still the ls process was killed
4) Wondering if the Lacie has Queued the request to delete the files, so now that I have repaired the disk it is locking up?

So nothing to lose, I will try and delete them from the standalone...

[fvdw]

If I try and list the files that I was attempting to delete when the Lacie died, the ls process is killed (I guess due to lack of memory)

it is remarkable that a command like "ls" is killed. I think that it doesn't use much memory as it reads the file tables and displays content of the table. Must be a damage in the fileystem or perhaps a listing of a file that no longer exists.

Did you try if you could read the content of those files ? But indeed trying to delete them again could not harm (I think)

[pfwaaa]

ls was dying in an Apple TimeMachine directory, so LOTS of files. I eventually used find to delete them all - which worked. Lacie still doesn't boot - it spins up all the disks and then sits there going :razz2
I still haven't got round to trying to find the Lacie's boot log to see what is going wrong.

So I have my files, but not sure how to get them off the Lacie for a complete reformat with possibly a swap to fvdw....