Hi Patrick, > > On Sun, Sep 26, 2010 at 01:30:48PM +0200, Michael Tautschnig wrote: > > Hmm, I thought that using the keyfile just meant "read key from this file" > > - can > > you just briefly confirm that this is not the case, i.e., using a keyfile is > > something totally different from using a passphrase? > > it indeed is. I couldn't find any documentation backing up my statement > but I just experimented with it a bit. Fact is with a test passphrase > in a file beeing loaded to a keyslot its not possible to unlock that > keyslot by using the passphrase in the file as a passphrase. > > Thats what will bite us here. > Hm, I've tried to create a patch but somehow I fail. Currently > I have come to a point where the luksFormat is ommitted for no > reason, making the following push commands fail.. really obscure. > Maybe you can have a look at it. >
[...] (useful patch) The problem with your patch most probably were the missing pre- and postconditions that push_command requires for later sorting the commands according to dependencies. But using the proper cryptsetup command lines from your patch I've tried to fix it: Could you re-test using 4.0~beta2+experimental18? Should now be available for download. Best, Michael
pgpp5ftzZEo00.pgp
Description: PGP signature
