Re: [gentoo-catalyst] Encrypted livecd's - need testers

[Erick M]

Nelson,

Ok, I understand, thanks for clarifying.
So we can narrow down all those scenarios to one type of attack: theft by
third part (or yourself).

It does not protect more the user while he uses it nor from potential
"after-use" trails. Either you lose the livecd along with your identity (or
data that leads to your identity) and you get caught or while using the
software you get caught (like your TOR connections have been detected).
The only purpose and advantage encryption would have is to obfuscate some
passwords like in the firefox example you gave.

Now, from a legal point of view, being caught with an encrypted material
whether livecd or not in major countries (UK,GER,FR,US,china) requires from
you the decryption key (us patriot act, uk RIP act, etc) or else you can
straight take up to few years in some cases without much chance of having of
good defense (china=torture?). So in 95% of cases you end up giving away
your key to prove that you are not a spy from whatever organisation and that
at least you hadn't that bad intention with your encrypted software. And you
do handle the key in the objective of lowering the sentence you get for
being caught in the first place.

I think that encryption has nothing to do with hiding. In the contrary, it
is like a big flag standing saying "hey look at me I got something to hide,
come and get me!". It is just obfuscating technology.

The real solution to your problem would be to use a steganographic layer (
http://en.wikipedia.org/wiki/Steganography ) . Not for the whole squashfs
but only for a single file (whatever the size) inside a clear livecd. Note
that 20% of the size of that file is really containing data, you do not want
to push too much (50%) or we get data loss (blocks from different containers
overwriting them) in an exponentially manner.

You want to be able to *deny* that you are in possession of such material.
Go from the basis that if you get caught you will *have to* handle your key
away. That is real practice because you can get 5 times more being secretive
than actual real sentence against the data you want to hide.
A steganographic FS will allow you when being caught with your livecd of
saying first: "it is a clear livecd!" Sounds idiotic but believe me, it is
the best start for the official police questioning. Then in the worst case
scenario, they find your single encrypted file and ask you for the key which
you will provide one of the many different you have set up (properties of a
steganographic FS), which will decrypt a part of that encrypted file,
discovering data that will not incriminate you so far for just having a
picture of your dog.
Charges are dropped, you justify your secretive attitude as being respectful
of your privacy rights and next morning you wake up in your bed!

Because I want to be fair, I think having an encryption layer is great for
catalyst, but when related to the specific purpose you described you would
better at least give a try to a steganographic FS if you really fear the
sentence you can get for the data you are hiding.

You will not find much (I mean actual real software) besides some
linux-2.2tweak over ext2 "proof-of-concept" (10years old not stable
unreliable) and
an update by some chinese with 2.4 but the whole is mainly broken and I
guess somehow a little taboo, the projects seems dead, no main other
projects have been replaced.

You can try an implementation I have worked on few years ago. It does
everything that I have described (in a non friendly C hardcore way) so far
and is called denyfs.

It is not a driver, and can be started in userland if the correct losetup
and cryptsetup have been done.

http://www.openchill.org/2005/06/denyfs_a_steganographic_file_s.php#more

have a look there, it is not fully stable, requires manual compilation and
configuration though it does the job (I made a quick GUI in gtk if you
provide the gtk USE flag). Follow the howto to get a grip on it. And
remember if you want to retrieve with a 90% probability your data as you
have put them in the box, do not exceed 15-20% of the total size of the
file! And even do not be surprised when it happens.

Steganography is a concept that aims at small and *static* file system. Do
not even think about putting an OS(where files are dynamically arranged
again and again) inside a steganographic FS, it is as of the concepts and
mathematics we have simply impossible.

I didn't realized I wrote so much, I'm just passioned by this topic because
of past experiences moving from one country to another.  I am currently
developing a Portage based GNU/Linux natively encrypted OS and I'm about to
re open DenyFS inside that distribution by stabilizing it, hence my reason
for being so communicative.

Thanks for reading

erick

On 7/1/07, Nelson Batalha <nelson_batalha@××××.pt> wrote:
>
> Hi Erick,
>
> There are many uses for this!
>
> They mainly come from the fact that now you can have sensitive information
> everywhere on your cd root, and not be afraid of losing your cd, either
> physically (happens to me all the time), or in the net if you don't want an
> open distribution.
>
> -Read on for examples:
>
> 1) If you're in a country like China and you can't have applications like
> Tor on your desktop (suspicious), you can just make a livecd and try to
> disguise it as something else by filling the filesystem. Also it's portable
> and replicable. You could also encrypt your hard drive, but this way you
> don't have to worry if they take it for testing. Specially if using luks on
> the desktop (no plausible deniability). It's also much easier to hide a
> mini-cd/dvd physically.
>
> 2) Also for instance, I'm going away next semester and I won't be taking a
> laptop. However I would like to use gentoo, my favourite programs and have
> my passwords stored in them (like Firefox), and transport some personal
> and/or sensitive files. (only option is put those files in an encrypted
> container and extract them on *every* boot).
>
> 3) If you're creating some official livecd and would like to test it with
> some group, but for security reasons you prefered if nobody else tested it.
>
> 4) In general companies/organizations can create a easily updatable
> portable working environment and mail it or publish it online.
>
> Etc.
>
> Take care,
> Nelson
>