On 11/01/2008 Robert Kelly wrote: > By grepping through a hexdump of the partition, I was able to find that > there was a LUKS header present. Since it seemed to be near the beginning of > the partition (short time to find it), I stored a hexdump of the partition > in a file and used vim to locate the header. From this, I determined that > the header appeared to be offset 3MB from the beginning of the partition, so > I used dd to copy it starting from that offset to another partition. This > was successful and all data was recovered. I am still not certain what is > responsible for this error, but I thought I would share my experience in > case it is helpful to someone else. Hey Robert, I'm part of the group that maintains cryptsetup in Debian: www.debian.org Back in July 2007 we got a bugreport which suggests to provide some 'repair tools' for cryptsetup because of several data corruption issues cryptsetup and/or the linux dm-crypt/device-mapper implementation has. You can find the relevant bug report at http://bugs.debian.org/432150 Now from what i read in your mails, it seems like you did exactly what the bug submitter suggested: salve or reconstruct the luks header. could you elaborate on how exactly you did that, maybe in a way that could be automated in a script? i already got that 'hexdump -C -n 1024 /dev/partition' gives information about the used cipher etc. thanks in advance, jonas --------------------------------------------------------------------- dm-crypt mailing list - http://www.saout.de/misc/dm-crypt/ To unsubscribe, e-mail: dm-crypt-unsubscribe@xxxxxxxx For additional commands, e-mail: dm-crypt-help@xxxxxxxx
