[ Date Index ] [ Thread Index ] [ <= Previous by date / thread ] [ Next by date / thread => ]
On 14/12/11 09:35, Tony Sumner wrote: > RESULT!! and thanks to everyone who helped. I can now mine the > partition and copy files to the new Debian system, which is of course > an ext3 fs. > > Tony Glad it all worked out Tony - you could have manually controlled the missing loopback devices with losetup but kpartx should have done that for you automatically. That's partly my fault, because I forgot to ask you to make sure you had multipath-tools and dmsetup installed as well (which as you didn't have LVM2 installed initially, you probably didn't). Also, I note you've played it safe with ext3 but ext4 really is production ready these days - you can do an in-place upgrade if necessary. Just stay away from reiserfs from now on :) You can also specify offsets in the original dd image file by bytes x sector and mount partitions from within it but you figured it out anyway: the important thing is you got your files back. Don't forget to check your recovered /var/log/* for clues as to why it went down (hint: "grep -i ata /var/log/dmesg*") and after you've nuked the failed disk make very, *very* sure it's ok before ever trusting it with data again. Personally I'd chuck it away after taking the magnets out for my fridge door. Moral of the story: when disks go bad, always immediately attempt to completely image them and work from the backup. Running smartctl/badblocks/RAID rebuilds or anything that isn't dd is just asking for trouble. Cheers, Mat -- The Mailing List for the Devon & Cornwall LUG http://mailman.dclug.org.uk/listinfo/list FAQ: http://www.dcglug.org.uk/listfaq