nanog mailing list archives

Re: Marriott wifi blocking


From: Jay Ashworth <jra () baylink com>
Date: Fri, 03 Oct 2014 23:32:39 -0400

Hugo, I still don't think that you have quite made it to the distinction that we are looking for here.

In the case of the hotel, we are talking about an access point that connects via 4G to a cellular carrier. An access 
point that attempts to create its own network for the subscribers devices. A network disjoint from the network provided 
by the hotel or its contractor.

This is a different case from the circumstance in a business office where equipment is deployed to prevent someone from 
walking in with an access point /which pretends to be part of the network which the office runs./

In the latter case, the security hardware is justified in deassociating people from the rogue access point, /because it 
is pretending to be part of a network it is not authorized to be part of/.

In the Marriott case, that is not the circumstance. The networks which the deauth probes are being aimed at are 
networks which are advertising themselves as being /separate from the network operated by the hotel/, and this is the 
distinction that makes Marriott's behavior is unacceptable. 

(In my opinion; I am NOT a lawyer. If following my advice breaks something, you get to keep both pieces.)

On October 3, 2014 11:04:08 PM EDT, Hugo Slabbert <hugo () slabnet com> wrote:
On Fri 2014-Oct-03 19:45:57 -0700, Michael Van Norman <mvn () ucla edu>
wrote:

On 10/3/14 7:25 PM, "Hugo Slabbert" <hugo () slabnet com> wrote:

On Fri 2014-Oct-03 17:21:08 -0700, Michael Van Norman <mvn () ucla edu>
wrote:

IANAL, but I believe they are.  State laws may also apply (e.g.
California
Code - Section 502).  In California, it is illegal to "knowingly and
without permission disrupts or causes the disruption of computer
services
or denies or causes the denial of computer services to an authorized
user
of a computer, computer system, or computer network."  Blocking
access to
somebody's personal hot spot most likely qualifies.

My guess would be that the hotel or other organizations using the
blocking tech would probably just say the users/admin of the rogue
APs
are not authorized users as setting up said AP would probably be in
contravention of the AUP of the hotel/org network.

They can say anything they want, it does not make it legal.

There's no such thing as a "rogue" AP in this context.  I can run an
access point almost anywhere I want (there are limits established by
the
FCC in some areas) and it does not matter who owns the land
underneath.
They have no authority to decide whether or not my access point is
"authorized."  They can certainly refuse to connect me to their wired
network; and they can disconnect me if they decide I am making
inappropriate use of their network -- but they have no legal authority
to
interfere with my wireless transmissions on my own network (be it my
personal hotspot, WiFi router, etc.).  FWIW, the same is true in
almost
all corporate environments as well.

Thanks; I think that's the distinction I was looking for here.  By 
spoofing deauth, the org is actively/knowingly participating on *my 
network* and causing harm to it without necessarily having proof that 
*my network* is in any way attached to *their network*.  The assumption

in the hotel case is likely that the WLANs of the "rogue" APs they're 
targeting are attached to their wired network and are attempts to
extend 
that wireless network without authorization (and that's probably 
generally a pretty safe assumption), but that doesn't forgive causing 
harm to that WLAN.  There's no reason they can't cut off the wired port

of the AP if it is connected to the org's network as that's their 
attachment point and their call, but spoofed deauth stuff does seem to 
be out of bounds.

I'm not clear on whether it runs afoul of FCC regs as it's not RF 
interference directly but rather an (ab)use of higher layer control 
mechanisms operating on that spectrum, but it probably does run afoul
of 
most "thou shalt not harm other networks" legislation like the 
California example.


/Mike



-- 
Hugo

-- 
Sent from my Android phone with K-9 Mail. Please excuse my brevity.


Current thread: