NAME

SYNOPSIS

DESCRIPTION

The gbde utility is the only official operation and management interface for the gbde(4) GEOM based disk encryption kernel facility. The interaction between the gbde utility and the kernel part is not a published interface.

The operational aspect consists of two subcommands: one to open and attach a device to the in-kernel cryptographic gbde module (attach), and one to close and detach a device (detach).

The management part allows initialization of the master key and lock sectors on a device (init), initialization and replacement of pass-phrases (setkey), and key invalidation (nuke) and blackening (destroy) functions.

The -l lockfile argument is used to supply the lock selector data. If no -l option is specified, the first sector is used for this purpose.

The -L new-lockfile argument specifies the lock selector file for the key initialized with the init subcommand or modified with the setkey subcommand.

The -n key argument can be used to specify to which of the four keys the operation applies. A value of 1 to 4 selects the specified key, a value of 0 (the default) means “this key” (i.e., the key used to gain access to the device) and a value of -1 means “all keys”.

The -f filename specifies an optional parameter file for use under initialization.

Alternatively, the -i option toggles an interactive mode where a template file with descriptions of the parameters can be interactively edited.

The -p pass-phrase argument specifies the pass-phrase used for opening the device. If not specified, the controlling terminal will be used to prompt the user for the pass-phrase. Be aware that using this option may expose the pass-phrase to other users who happen to run ps(1) or similar while the command is running.

The -P new-pass-phrase argument can be used to specify the new pass-phrase to the init and setkey subcommands. If not specified, the user is prompted for the new pass-phrase on the controlling terminal. Be aware that using this option may expose the pass-phrase to other users who happen to run ps(1) or similar while the command is running.

The -k keyfile argument specifies a key file to be used in combination with the pass-phrase (whether the pass-phrase is specified on the command line or entered from the terminal) for opening the device. The device will only be opened if the contents of the key file and the pass-phrase are both correct.

The -K new-keyfile argument can be used to specify a new key file to the init and setkey subcommands. If not specified, no key file will be used (even if one was previously used).

EXAMPLES

To attach an encrypted device:

The encrypted device has the suffix .bde so a typical command to create and mount a file system would be:

To detach an encrypted device:

Please notice that detaching an encrypted device corresponds to physically removing it, do not forget to unmount the file system first.

To initialize the second key using a detached lockfile and a trivial pass-phrase:

To invalidate your own masterkey:

This will overwrite your masterkey sector with zeros, and results in a diagnostic if you try to use the key again. You can also destroy the other three copies of the masterkey with the -n argument.

You can also invalidate your masterkey without leaving a tell-tale sector full of zeros:

This will overwrite the information fields in your masterkey sector, encrypt it and write it back. You get a (different) diagnostic if you try to use it.

SEE ALSO

HISTORY

AUTHORS

BUGS