i've got an XFS partition, it doesn't mount 'cuz the superblock can't be found, which was weird. i tried xfs_repair but it didn't give me any good results. i tried with xfs_repair -o assume_xfs /dev/sda and it still doesn't seem to work. so how can i mount this partition when it can't find the superblock? i know there are contents 'cuz i transfered files to it and umount it before rebooting. what happened between working and not working? sandeen, i transfered files to it, umount it, then reboot ktwilight, how big is the partition ktwilight, what kernel, which partition, what type of block device, etc 2.6.18-5, it's the entire hdd, /dev/hdb what do you mean by block device? that's enough what does file -s /dev/sdb say er hdb :) /dev/hdb: SGI XFS filesystem data (blksz 65536, inosz 256, v2 dirs) it should have a /dev/hdb1 huh, ok, I thought maybe something overwrote it it's umount, and i try my best not to do anything to do :) nothing wrong with using the whole bdev hm, but i can't mount either hdb or hdb1 which one *should* it be what used to mount :) i.e .what's in fstab nothing in fstab but i used hdb1 to mount previously what does file -s /dev/hdb1 say then? or fdisk -l /dev/sdb? /dev/hdb1: ERROR: cannot open `/dev/hdb1' (No such file or directory) Disk /dev/hdb: 250.0 GB, 250059350016 bytes 255 heads, 63 sectors/track, 30401 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Disk /dev/hdb doesn't contain a valid partition table the last sentence is quite... :/ er, so, you no longer have a partition table, but rather it is an xfs filesystem on the whole /dev/hdb... which is fine, but on the last boot you mounted hdb1? I'm having a hard time believing that nothing happened except file copies and a reboot :) looks to me like you mkfs'd /dev/hdb can you mount /dev/hdb? can't be... nope, i can't mount it what does dmesg say when you try? ktwilight: hexdump -C the first 512 bytes for us and use paste.pocoo.org a) you have no partition table, thus no /dev/hdb1, and b) "file" finds an XFS superblock directly on hdb, so... dmesg doesn't say anything, but mounting gives me "mount: Function not implemented" * sandeen thinks xfs should store mkfs time for cases like this :) sandeen: hehe lsmod | grep xfs ? :) um, so hexdump -C /dev/hdb? ktwilight: dd ... count=1 dd if=/dev/hdb bs=512 count=1 | hexdump -C xfs is loaded, i ave other xfs partitions mounted http://rafb.net/p/WLxJzd90.html I have a hard time believing that mount /dev/hdb /some/where fails but generates no kernel messages try mount -t xfs? http://rafb.net/p/ghKCgI36.html xfs_db -r -c "sb 0" -c p /dev/sdb and mount -t xfs /dev/hdb /some/where; dmesg | tail -n 10 http://rafb.net/p/GL95be48.html sorry s/sdb/hdb/ :) np ;) ew http://rafb.net/p/AcilLz30.html not a good ew i guess that segfault isn't nice errr XFS mounting filesystem sdb1 Ending clean XFS mount for filesystem: sdb1 oh sdb geesh different ;) you'd think I'd have this straight by now :) *sigh* it's hdb, NOT sdb though i have an sdb and it's working perfectly. let's not touch it :) so dmesg doesn't say anything when trying to mount hdb weird that xfs_db segfaults is this latest xfsprogs? 2.8.11-1 debian etch. it's "stable" :) might try very latest xfsprogs for fun ... just to see if we can examine it more so i should grab it off sid? tell me what you know so far or its assumptions or use xfs_metadump to get an fs image, and provide it to the sgi guys at a minimum xfs_db should not segfault hm well, so far, bare hdb has at least the beginning of an xfs filesystem signature, XFSB is supe block magic and, you have no partition table on hdb so basically i have no files on hdb? no way of recovering? not sure yet great :) but, I don't quite believe that you had /dev/hdb1 mounted, copied files, umounted, and suddenly you had no more partition table something else happened in between whether you did it yourself or something did it to you I think :) i don't believe it either, but that's the truth. well, rather than copy, it was a move. does that make any difference? :P what kind of partition table did you have? dos or gpt? dos your root inode nr from the hexdump is a little funky bad funky? k, finally gettin' sid's xfsprogs great, xfs_db segfaults too :) xfsprogs 2.9.4-2 hmm other same messages? cache_node stuff or no not really http://rafb.net/p/e4uYNC58.html same stuff o ok :| node is different. that helps? haha sandeen, would xfs_fsr help? no you have to mount it first :) :( how 'bout xfs_check or xfs_admin? not that it'd help anyway hm so what are my options? if xfs_db can't even recognize the filesystem things aren't so good would dd help? let's say i skip the first few bytes then mount it as an iso. no backup? hannesd, yes, but not that up to date :/ if i can recover, it'd be better. just for fun, does dd if=/dev/hdb bs=512 count=32 | hexdump -C | grep XFSB turn up more than 1 XFSB? * sandeen guesses at the count=32, but hdb1 probably started before 16k http://rafb.net/p/tHqWyb88.html i only see one. nope hm, so it's not good. so should i just forget about it? or is there a glimpse of hope somewhere? :) was gonna look over the first hexdump a bit ok 00 00 00 00 00 00 09 00 should be your root inode, which is not 22, so a little confused about why it is ssaying rootino 22 when you try xfs_db unless that's EINVAL that got lost can you use xfs_metadump to make an fs image? maybe the sgi guys can help further some monster ate it i see. how big would the fs image be? as big as the partition? at least bnaujok would probably like to see what's going on with xfs_db no it only dumps metadata no data so it should be significantly smaller hm but if db can't read it, metadump might not either ya gotta see this http://rafb.net/p/ojVANy73.html yeah, libxfs problem, I was sorta afraid of that well. underlying problem is what's on your disk, but then libxfs isn't coping well either maybe the first few K of the fs would be enough for them to dig in further hm I should be working on another fs right now ;-) :) k, so i will send xfs_metadump, and xfs_db to sgi? what else should i include in the email? the chat log maybe :) :) a heck load of stuff then email? support@xxxxxxx? xfs@xxxxxxx open source xfs list hexdump of hdb would be a place to start, along w/ xfs_db results nm, found. xfs@xxxxxxxxxxx and a sworn statement that oh really this used to be hdb1 but it magically moved :)