Hi Rafał, I have split your bug report because it includes two distinct issues.
1. Problem mounting loop-AES loop (#394406) 2. System left without /bin/mount after upgrade (#394457) On Sat, Oct 21, 2006 at 06:01:07AM +0200, [EMAIL PROTECTED] wrote: > It fails to mount a crypto partition (loop AES) - dispalys either the > can't find file system (see dmesg) standard message, or some bad > IO_CONTROLL like error (I think it was the case when I had to use the > /bin/mount.orig) Need more details: What is the exact error message? Please quote the error message in full and include the command you used. > I am a bit confused - should I now use 'cryptosetup' to access the > partition created by loopaes like: You don't need to use 'cryptsetup' (I think you mean 'cryptsetup' as no 'cryptosetup' exists in Debian?). That package is unrelated to loop-AES and the setup you describe is possible to use with just loop-aes-utils. > losetup -T -e aes-128 /dev/loop6 -C 200 -Sfoo /dev/hda12 > with fstab entry: > /dev/hda12 /cr/store jfs > user,noatime,nodev,nosuid,encryption=aes-128,itercountk=200,pseed=foo 0 0 OK. > Even if yes, then this change should be made more transparent, > otherwise users like me end up suddenly with system that fails to > auto mount some of partitions, or even without (working) "mount" > command at all - therefore I report it as grave, in my case it > rendered the PC unbootable (untill maintaince and fixing). There was no such change. If it stopped working, something else must have changed. We can probably find out if you send the above detailed error message. > When I by hand setup the losetup loop with identical options > and then call mount on /dev/loopX - it works fine as before. > Also, right after doing apt-get upgrade which upgraded the > loop-aes-utils probably (among other tools), the systeam was left > without ANY mount at all! there was no file /bin/mount, only file > /bin/moun.orig which I used to mount partitions and fixed system to > usable state. (This is #394457; I will reply in another message) > I tried downgrading loop-aes-utils to sarge version but it seems it > didnt helped. > After downgrade: > blackelf:/home/raf256# sha1sum `which mount` > 9cc043710d1287c0e79da9931c7a73c8c28feb07 /bin/mount > > before downgrade (in current testing 2.12r-13) > blackelf:/home/raf256# sha1sum `which mount` > e458b5155d717f00db7ed39373e1edff12042207 /bin/mount OK. This matches sha1sum of the binary in 2.12r-13. > it do not work in either version: > > blackelf:/home/raf256# mount /cr/store/ > Password: > mount: wrong fs type, bad option, bad superblock on /dev/loop0, > missing codepage or other error > In some cases useful info is found in syslog - try > dmesg | tail or so > > (yes, ofcourse I am using the right mountpoint-partition, giving > correct password, and there are correct options in /etc/fstab I > doublechecked all) I have tried to reproduce the problem using the exact same losetup command and fstab options. This worked without problems using both loop-aes-utils 2.12r-13 (testing) and 2.12p-4sarge1 (stable). Which version of the loop-AES kernel modules are you using? Can you send the output of "/sbin/lsmod | grep loop" after mount failed? Perhaps the problem is related to module autoloading, so please send lsmod at the time after mount, but before manual losetup. cheers, Max
