Re: [pfSense-discussion] pfSense comment packetpushers.net

2011-05-25 Thread Ian Bowers
I think the gist of what he's saying is that because it's running on a
*nix, anyone can log in and install any software they want on it.
Ultimately this is a gaping security hole from certain perspectives.
I don't mean that the firewall software or the OS contains gaping
security holes.  Don't get me wrong, I love OpenBSD, pf, FreeBSD, and
PFsense when I tried it.  What Greg is saying is that because, in this
case, it's FreeBSD underneath, anyone with root access can go in and
install stuff.  So the only way you can certify the performance and
security is as it exists when its still in the box.  Then take an ASA
for example.  You get it in state X.  It's capable of almost limitless
config variations, but the underlying functions the platform can
perform are static.  You can never SSH from the ASA to another device.
 you can never run mysql on it.  And all I mean by this is that some
asshole or rogue IT guy can come along and install whatever they want
on a PFSense firewall.  In a proper environment there would be
controls against this, but thats dependent on the environment the
device is installed in so you can't really roll that up into a
security specification/certification.  I think he's also getting at
that it's just software, and it depends on the hardware you run it on.
 Take Soekris for example...  Love Soekris, love their hardware, but I
hate VIA chipsets.  Less now as before, but over time they've proven a
headache and a burden.  You can't certify pfsense to perform and
operate a certain way unless you wrap up the software with specific
tested hardware.  and having the ability to install arbitrary software
on it makes it open to more than just config errors.

I'm digressing a little bit, but it's mostly related.  Basically his
point is you can't trust IT staff to not muck something up.  So having
a platform where arbitrary stuff can be installed isn't something that
can be afforded in many cases.

Again I'm a huge proponent of open source, BSD, and pf.  And
personally believe they're a great solution in many of cases.  I'm
just responding based on what I think Greg's thinking.  He's very
knowledgeable and he's been in the networking game a while.  I've
rarely seen him hate on products simply because they're niche.

-Ian

On Wed, May 25, 2011 at 11:59 AM, BSDwiz bsd...@gmail.com wrote:

 Guys,
 I was Listening to a packetpushers.netpodcast regarding the topic of
 firewalls and decided to chime in. I thought you may have some thoughts or
 opinions to add. Basically, I mentioned pfSense and was not very happy with
 his(Greg Ferro) response.  If you get a minute, check out this guys
 reasoning behind not using pfSense.
 http://packetpushers.net/show-42-hating-firewalls-wrong-checkpoint/#comment-425

 Best,
 Phil(phospher)



-
To unsubscribe, e-mail: discussion-unsubscr...@pfsense.com
For additional commands, e-mail: discussion-h...@pfsense.com

Commercial support available - https://portal.pfsense.org



Re: [pfSense-discussion] pfSense comment packetpushers.net

2011-05-25 Thread Tim Dressel
This sort of points the finger then at a commercial need for a hardened
pfsense product running on a specialized ASIC of some sort.

So when can Chris sort that out? :)

On Wed, May 25, 2011 at 9:32 AM, Ian Bowers iggd...@gmail.com wrote:

 I think the gist of what he's saying is that because it's running on a
 *nix, anyone can log in and install any software they want on it.
 Ultimately this is a gaping security hole from certain perspectives.
 I don't mean that the firewall software or the OS contains gaping
 security holes.  Don't get me wrong, I love OpenBSD, pf, FreeBSD, and
 PFsense when I tried it.  What Greg is saying is that because, in this
 case, it's FreeBSD underneath, anyone with root access can go in and
 install stuff.  So the only way you can certify the performance and
 security is as it exists when its still in the box.  Then take an ASA
 for example.  You get it in state X.  It's capable of almost limitless
 config variations, but the underlying functions the platform can
 perform are static.  You can never SSH from the ASA to another device.
  you can never run mysql on it.  And all I mean by this is that some
 asshole or rogue IT guy can come along and install whatever they want
 on a PFSense firewall.  In a proper environment there would be
 controls against this, but thats dependent on the environment the
 device is installed in so you can't really roll that up into a
 security specification/certification.  I think he's also getting at
 that it's just software, and it depends on the hardware you run it on.
  Take Soekris for example...  Love Soekris, love their hardware, but I
 hate VIA chipsets.  Less now as before, but over time they've proven a
 headache and a burden.  You can't certify pfsense to perform and
 operate a certain way unless you wrap up the software with specific
 tested hardware.  and having the ability to install arbitrary software
 on it makes it open to more than just config errors.

 I'm digressing a little bit, but it's mostly related.  Basically his
 point is you can't trust IT staff to not muck something up.  So having
 a platform where arbitrary stuff can be installed isn't something that
 can be afforded in many cases.

 Again I'm a huge proponent of open source, BSD, and pf.  And
 personally believe they're a great solution in many of cases.  I'm
 just responding based on what I think Greg's thinking.  He's very
 knowledgeable and he's been in the networking game a while.  I've
 rarely seen him hate on products simply because they're niche.

 -Ian

 On Wed, May 25, 2011 at 11:59 AM, BSDwiz bsd...@gmail.com wrote:
 
  Guys,
  I was Listening to a packetpushers.netpodcast regarding the topic of
  firewalls and decided to chime in. I thought you may have some thoughts
 or
  opinions to add. Basically, I mentioned pfSense and was not very happy
 with
  his(Greg Ferro) response.  If you get a minute, check out this guys
  reasoning behind not using pfSense.
 
 http://packetpushers.net/show-42-hating-firewalls-wrong-checkpoint/#comment-425
 
  Best,
  Phil(phospher)
 
 

 -
 To unsubscribe, e-mail: discussion-unsubscr...@pfsense.com
 For additional commands, e-mail: discussion-h...@pfsense.com

 Commercial support available - https://portal.pfsense.org




RE: [pfSense-discussion] pfSense comment packetpushers.net

2011-05-25 Thread Greg Hennessy
Doesn’t seem to be unreasonable TBH.  It's a case of horses for courses.
Some use cases take seperation of duties really seriously.

Can completely understand where he is coming from.
The commentary on Chokepoint is particularly apt.


Greg

From: BSDwiz [mailto:bsd...@gmail.com]
Sent: 26 May 2011 2:00 AM
To: discussion@pfsense.com
Subject: [pfSense-discussion] pfSense comment packetpushers.net


Guys,
I was Listening to a packetpushers.nethttp://packetpushers.netpodcast 
regarding the topic of firewalls and decided to chime in. I thought you may 
have some thoughts or opinions to add. Basically, I mentioned pfSense and was 
not very happy with his(Greg Ferro) response.  If you get a minute, check out 
this guys reasoning behind not using pfSense.
http://packetpushers.net/show-42-hating-firewalls-wrong-checkpoint/#comment-425

Best,
Phil(phospher)