Re: Security Comparisons

2007-11-10 Thread knitti
On 11/10/07, Douglas A. Tutty [EMAIL PROTECTED] wrote:

 of philosophy.  Linux is about making all kinds of toys work in a
 hot-plug way and allow people to boast about their uptime.  OpenBSD is
 about security.

I would add usability (conciseness, least surprise and coherency) and
thus maintainability to the list. I end up having less to do for OpenBSD
Servers to keep them happy running than for some Debian boxes, and
Debian _is_ damn well maintainable.

--knitti



Re: Security Comparisons

2007-11-10 Thread Douglas A. Tutty
On Sat, Nov 10, 2007 at 05:52:03PM +0100, knitti wrote:
 On 11/10/07, Douglas A. Tutty [EMAIL PROTECTED] wrote:
 
  of philosophy.  Linux is about making all kinds of toys work in a
  hot-plug way and allow people to boast about their uptime.  OpenBSD is
  about security.
 
 I would add usability (conciseness, least surprise and coherency) and
 thus maintainability to the list. I end up having less to do for OpenBSD
 Servers to keep them happy running than for some Debian boxes, and
 Debian _is_ damn well maintainable.
 

True.  Although I have't run into the problem, there have been many
cries for help on the debian-user list when udev renames drive devices
and a machine refuses to boot.  The answer has been to use linux's
ability to mount by LABEL= instead of /dev/* so that the kernel looks at
all the devices for the approprate filesystem label.  Just an example of
a surprise gotcha.

Doug.



Re: Security Comparisons

2007-11-09 Thread Darren Spruell
On Nov 9, 2007 10:53 AM, new_guy [EMAIL PROTECTED] wrote:
 If this is off-topic, I apologize. Just tell me and I'll go away ;)

 I'm having discussions with a coworkers about moving to OpenBSD for
 Apache/PHP web hosting. Right now, we use various Linux distros. I have no
 problem with that. Linux is cool... but it's takes more time to secure and
 manage. I like the Suhosin (Hardened PHP patch in OpenBSD's PHP package) and
 the fact that Apache is chrooted by default. We even uploaded some php
 exploit code onto a test OpenBSD box (r57shell) to see how well it contained
 the exploit. It worked well. All of these demos and discussions are
 informal. So here's the question: Are there any formal/corporate comparisons
 that demonstrate the enhanced security of OpenBSD when compared to other
 solutions in this space that we can provide to upper management?

Sadly, justifying the obvious through these means is often a requirement.

Here's an approach you might consider. Take a best practice /
standards guide such as from NIST:

http://www.itl.nist.gov/lab/bulletns/bltndec02.htm
http://csrc.nist.gov/publications/drafts/800-44-Version2/Draft-SP800-44v2.pdf

And for the points your organization feels are important (like what
you've listed above), map how OpenBSD's implementation and OS approach
addresses those points. You'll find this is a pretty good indicator
and should be well accepted by the folks that matter.

DS



Re: Security Comparisons

2007-11-09 Thread new_guy
Darren Spruell wrote:
 
 
 Sadly, justifying the obvious through these means is often a requirement.
 
 Here's an approach you might consider. Take a best practice /
 standards guide such as from NIST:
 
 http://www.itl.nist.gov/lab/bulletns/bltndec02.htm
 http://csrc.nist.gov/publications/drafts/800-44-Version2/Draft-SP800-44v2.pdf
 
 And for the points your organization feels are important (like what
 you've listed above), map how OpenBSD's implementation and OS approach
 addresses those points.
 

Thanks... that's a good suggestion. I found the Secunia OS advisories very
telling as well. Comparing OpenBSD 3.x (85 Advisories) to Debian 3.x (577). 

http://secunia.com/product/
-- 
View this message in context: 
http://www.nabble.com/Security-Comparisons-tf4779123.html#a13676309
Sent from the openbsd user - misc mailing list archive at Nabble.com.



Re: Security Comparisons

2007-11-09 Thread Douglas A. Tutty
On Fri, Nov 09, 2007 at 02:27:16PM -0800, new_guy wrote:
 Darren Spruell wrote:
  
  
  Sadly, justifying the obvious through these means is often a requirement.
  
  Here's an approach you might consider. Take a best practice /
  standards guide such as from NIST:
  
  http://www.itl.nist.gov/lab/bulletns/bltndec02.htm
  http://csrc.nist.gov/publications/drafts/800-44-Version2/Draft-SP800-44v2.pdf
  
  And for the points your organization feels are important (like what
  you've listed above), map how OpenBSD's implementation and OS approach
  addresses those points.
  
 
 Thanks... that's a good suggestion. I found the Secunia OS advisories very
 telling as well. Comparing OpenBSD 3.x (85 Advisories) to Debian 3.x (577). 
 
 http://secunia.com/product/

However, you should read their PLEASE NOTE: comment.  Especially when
you figure that the reports for Debian are for all the packages in
debian (thousands of them) whereas OpenBSD doesn't have as many pieces.
They specifically say not to use the number of advisories to compare the
relative security of the products on which they report.

You also have to look at the duration of support.  OpenBSD comes out
with a new version every six months.  Debian comes out every few years.
Since Debian is designed with continuous updates possible, the only
impitus for a new OS version is new versions of software.  Otherwise,
the Debian security team takes security advisories in newer versions and
backports them to the version supplied in the current stable branch.  

If you look specifically at, for example, Debian 3.1 (Sarge) and want to
compare it with OpenBSD, you'd have to look at the dates from Sarge
release to Etch (4.0) release and count the security advisories (which
are both security and important bug fixes).  Then look at the security
advisories for OpenBSD in that time.  Then weed out of Debian's count
those updates that applied to applications that aren't in OpenBSD, and
weed out bugfixes only (that may have been applied to OpenBSD -current
but not backported in to -patch).  

The one thing you will find is that there have been more updates to any
single version of the Linux kernel than to the OpenBSD kernel.  Its the
nature of the beast: Linux is all about new features to work on new
hardware.  To me the biggest difference between Linux and OpenBSD is one
of philosophy.  Linux is about making all kinds of toys work in a
hot-plug way and allow people to boast about their uptime.  OpenBSD is
about security.  If you add a new piece of hardware, do a reboot and
forget about uptime as a quality indicator.  Its not a fair comparision
since OpenBSD handles USB stuff too but the philosophical difference is
there.

Doug.