On Thu, 10 Nov 2005, Spencer Dawkins wrote:
Just to be clear - is the problem "ad hoc mode" or "ad hoc mode with SSID
ietf"?
The problem basically works out to something like this...
A host with the magic settings, or defaults comes up, for whatever reason
it can't associate with an accesspoint, so it says hey: I'll flop over to
bss mode and become an adhoc node.
Now another host comes along as says, hey the strongest network I see is
this adhoc network, so I'll join that.
So you have a bunch of hosts participating in this bss network, and
because they're not being managed by an ibss node (an ap) their beacons
and any traffic they send clobber traffic from the ap's around them making
the situation worse.
Now, the ibss ap's have their power output turned down so that they don't
clobber each other, because a certain ap density is needed support all the
users on this network. So if the ap is transmitting at 15mw and you
have a laptop with 100mw card which One wins?
The problem with a node that's decided to become adhoc is when would it
decide to change back? It won't. Probably when your fiddle with your card
settings, reset the card, sleep the laptop or reboot, that will be enough.
Certain implentations, eg macosX 10.1 would switch to an adhoc network
with the same ssid as the mananged netowrk even if they were configured
only to connect to managed networks.
So, good hygiene is:
Configure your laptop to stick to the ssid ietf64
Configure you card to only operate in managed, ibss or accesspoint
networks.
If you have a card with selectable output power (like an old cisco, prism
2, or atheros) pick something below 100mw like 15mw or or 30mw.
if you have a card with a density setting like and old lucent orinoco
card, set it to high.
if you have 802.11a support use it.
The last time we were in Minneapolis, Dean Willis noticed that the wireless
projector controls in the conference rooms used 802.11b ad hoc ... in an
increasingly IP-deviced world, if the problem is "ad hoc mode", we are going
to die at an increasing rate over time.
The number of devices in your pocket and in the environment with radios
does indeed increase over time. Couple that with the challenges of working
in a new space, with days or hours of setup time, no decent simulation
tools for a room with 100 tons of meat and 800 radio's in it. and host
implementations of widely varying quality, and you have a challenging
dynamic environment that should make every host scared.
Spencer
_______________________________________________
Ietf@xxxxxxxx
https://www1.ietf.org/mailman/listinfo/ietf
--
--------------------------------------------------------------------------
Joel Jaeggli Unix Consulting joelja@xxxxxxxxxxxxxxxxxxxx
GPG Key Fingerprint: 5C6E 0104 BAF0 40B0 5BD3 C38B F000 35AB B67F 56B2
_______________________________________________
Ietf@xxxxxxxx
https://www1.ietf.org/mailman/listinfo/ietf