Mensaje archivado #3295 de la Lista os2-wireless_users@2rosenthals.com

De: Lewis G Rosenthal <os2-wireless_users@2rosenthals.com> Encabezados Completos
Mensaje no decodificado
Enviador: os2-wireless_users-owner <os2-wireless_users-owner@2rosenthals.com>
Asunto: [OS2Wireless] OS/2 Wireless FAQ
Fecha: Tue, 21 Dec 2004 23:20:26 -0500
Para: os2-wireless_users@2rosenthals.com

(Just responding in reverse order to the follow-up posts in this thread...)

On 12/21/2004 03:05 pm, Christian Langanke thus wrote :

Lewis,

great collection of information!

Thanks! It's amazing what's gotten stuck between my ears since all of this wireless stuff hit the scene!! :-)


For now I just som minor points, I have to think about more questions later on:

- AFAIK the "24-bit header" of WEP is called the initialization vector, which should be a random number forr each frame. One of the biggest vulnerabilities of the WEP protocol is that many vendors don't implemment that as a random number, but just count up from zero

Right on both counts, and I will make a note to correct my verbage concerning the IV. The main vulnerability of WEP, of course, is the fact that the keys repeat. Sniff enough packets over a long enough period of time (the busier the WLAN, the shorter the necessary time), and one can easily discern the keys being used.


- Concering the dynamic 24-bit part and the static part please add a point about "open Systems" and "shared Key Systems", this topic is also referred to "authentication". If set to"shared key" on behalf of the access point,  the client would need to  authenticate itself, unfortunately this is done with the static parts of the keys only (!!!),  without any initialization vector. The drawback of this is that alone this authetication makes the WEP method even more unsafe and easier to crack, so it strangely makes a WEP secured access point more safe when this authetication is turned of or set to "open system".

Good point, again. I'll add this in.

- as stated in my documentation, please add something like "always use encrpytion when accessing private WLans" and "better use 64-bit or _any_ WEP encryption than none". The user should be told to change the keys periodically. The more data is sent over WEP encrypted WLAN, the quicker it can be cracked. With medium to heavy throughput WEP64 can be cracked after some hours, WEP128 may need some days. For an average user, just surfing a bit in the evening, as a rule of thumb I would suggest a week for a change of WEP128 keys, and half a week for WEP64. I don't do that myself, and I am almost certain that nobody does that, but it is important to know that it is _unsafe_ not to do so. Furthermore, in Germany it is at least illegal to get into a WEP secured private WLAN. If that is the case, it may even make sense to set any key and never change it anymore...

Per my comments, above. I'll add all of this in the WEP-related section of the FAQ.

- topic 6: there is a term of "infrastructure mode" and at least another one.I think this would fit in here

BSS. Yes, I'll make the distinction between BSS, IBSS, and AdHoc.

- topic 7: I would turn the question around: Is an access point a router ?

Hmmm... Okay. (For anyone else following this exchange, the following may get a bit sticky; please bear with us, as Christian and I have a little fun...). WDS (Wireless Distribution System) mode can actually put the WLAN on a different network from the wired LAN, thus making an AP "route" packets... So, perhaps, we have two different questions to answer, eh? :-)

Which reminds me, related to the AP section, I should probably mention something about bridging and WDS.

- topic 8: SSID is AFAIK also referred to as System Service Identifier

Hmmm... I don't think I'd heard that one, but I'll look it up (not doubting you; just curious to see what else I may have missed).

- topic 10: please include that with certain WLAN sniffer software (either Win32 or linux ?) the SSID can still be spyed out even if broadcast is turned off. IMHO this is very important to know, else users would get a false idea of security they would establish. Nevertheless it is still a good idea to turn off  SSID broadcast for a private WLAN AP, to make it not too easy for hackers.

Good idea. I should have made it clear that XP isn't the only way to snatch an SSID out of the air when beaconing is turned off.

- Please include a section on MAC address filtering. Unfortunately also this can be spyed out by software like stated above for SSID, but also this should be enabled. It makes it at least not possible for a drive-by surfer to instantly get a connection. Isntead, he would need to spy out a MAC address, and come back another time to wait until this is not active while the AP is active, so that he can use this MAC addess.

Yes, I did make a quick note in the security section that I want to add a mention of MAC filtering.

Thanks for the tips. I'll work them in tomorrow night.

--
Lewis
------------------------------------------------------------
Lewis G Rosenthal, CNA Rosenthal & Rosenthal, LLC
Accountants / Network Consultants  New York / Northern Virginia           www.2rosenthals.com
eComStation Consultants                  www.ecomstation.com
Novell Users International        www.novell.com/linux/truth
------------------------------------------------------------ =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

To unsubscribe from this list, send a message to
steward@2rosenthals.com with the command
"unsubscribe os2-wireless_users" in the body
(omit the quotes).

For help with other commands, send a message
to steward@2rosenthals.com with the command
"help" in the body (omit the quotes).

This list is hosted by Rosenthal & Rosenthal
P.O. Box 281, Deer Park, NY 11729-0281. Non-
electronic communications related to content
contained in these messages should be directed
to the above address. (CAN-SPAM Act of 2003)

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=




Suscribirse: Todos, Compendio, Indice.
Desuscribirse
Correo al dueño de la Lista