[Pkg-cryptsetup-devel] Bug#371135: About Bug#371135: suggestion

[Jonas Meurer]

On 17/06/2006 Michael Gebetsroither wrote:
> >  With changing keys there is absolutely no way to identify what is
> >valid swap space area. I see two possible approaches that _might_ be
> >_theoretically_ doable:
> 
> creativity ;)
> i think there are a few more choices.
> 
> >  1. If crypttab defines an encrypted swap, use vol_id check for both
> >     the created mapping AND the actual device.
> >
> >  2. Use Jari Ruusu's watermark attack and explicitly disallow ESSIV
> >     encryption mode for swap.
> 
> 3. deprecate the usage of plain cryptodevices for swap and use luks instead.
>    we have to explicitly format on every reboot, but we would be able  
> to check.
>    This does not help against user failuers, where another luks partition is
>    given as swap partition.

first, we should not require swap partitions to be LUKS encrypted. plain
dm-crypt also has advantages.
second, i don't think that we should automatically LUKS-format a partition
at any time.

> 4. map the swap with an offset and write a magic cookie in the first 
> sector(s)
>    maybe also left sector(s) at the end of the device, because the  
> first sector
>    get's easily destroyed by a live-cd or hibernation.

then the encrypted swap partitions may always be identied, and i don't
like that idea too.

> 5. if we force luks for swap we could use the UUID to check if it's really 
> our
>    device. beside this we should add a command to cryptsetup to actually set
>    the UUID or to reinitialise the master key (i'm more comfortable with
>    reinitialise the master key, suggestions?)

that may be an interesting feature, but ...

> plain cryptsetup should imho be deprecated for swap.

i object here. again, forcing the admin to use LUKS for swap is not what
we want.

> >  So the two practical approaches that I see, are: allow users to hang
> >themselves, or do not allow automatically used encrypted swap at all.
> 
> linux _HAS_ to do as the user said.

exactly, and i don't see why we should take the responsibility that
belongs to the admin. i like the idea of testing for common situations
where data IS INDEED DESTROYED, but not to build further barriers for
the admin.

> >  Sure, you could add yet another test for partition size and require
> >that for encrypted swap the size of the swap is given. This doesn't
> >sound too practical, and still does not provide absolute protection
> >against human errors - I could have 4G swap and 4G encrypted /usr/local.
> >In which case the issues of case 1 are present again. You can't use
> >partition type either, as swap could be a file.
> 
> That would be another possibility. put the size (in blocks) in the 
> configfile.

again another barrier that doesn't make the whole system more secure
against data loss.

by the way, great to see you on the debian cryptsetup list, Michael.

...
 jonas