1995-10-29 - Re: New release of CFS Unix encrypting file system available

Header Data

From: Matt Blaze <mab@crypto.com>
To: cypherpunks@toad.com
Message Hash: b3f1d57758f84eeaa324b347c355238e616c9b97f6e44193a0f82565a4f3edae
Message ID: <199510282357.XAA09892@crypto.com>
Reply To: <199510282012.NAA25761@infinity.c2.org>
UTC Datetime: 1995-10-29 00:07:10 UTC
Raw Date: Sun, 29 Oct 1995 08:07:10 +0800

Raw message

From: Matt Blaze <mab@crypto.com>
Date: Sun, 29 Oct 1995 08:07:10 +0800
To: cypherpunks@toad.com
Subject: Re: New release of CFS Unix encrypting file system available
In-Reply-To: <199510282012.NAA25761@infinity.c2.org>
Message-ID: <199510282357.XAA09892@crypto.com>
MIME-Version: 1.0
Content-Type: text/plain


Anonymous writes:
> >
> >What happens to hard links?
> >
> >mkdir foo bar
> >CFS_set_directory_key -directory ./foo -key foo-key
> >CFS_set_directory_key -directory ./bar -key bar-key
> >cp /etc/passwd ./foo/test1
> >ln ./foo/footest ./bar/bartest
> >cmp ./foo/footest ./bar/bartest
> 
> This is a serious flaw. The emperor has no clothes. People should
> sue at&t for this shit.

I'm not sure why I'm bothering to respond to this, but I'd hate to
think someone might take the above message seriously and think that
there's some kind of "serious flaw" in CFS demonstrated by this sequence
of (hypothetical, incorrect) commands.  So here goes:

What on earth are you talking about?

As I pointed out in a previous message, that's not how CFS works - you
can't link across encrypted directories.

There may be (and probably are) bugs in or attacks against CFS, but this
isn't one of them.

-matt






Thread