Hi Jason,
Just to provide a bit more detail to Yeoh's response:
To support mesh security in hardware, your wireless card needs to
support multiple encryption keys and management frame encryption. The
driver advertises this capability to the 802.11 stack via the flags:
IEEE80211_HW_MFP_CAPABLE and IEEE80211_HW_SUPPORTS_PER_STA_GTK. The
rt2800 driver does not seem to support these:
in rt28000lib.c:rt2800_probe_hw_mode()
rt2x00dev->hw->flags =
IEEE80211_HW_SIGNAL_DBM |
IEEE80211_HW_SUPPORTS_PS |
IEEE80211_HW_PS_NULLFUNC_STACK |
IEEE80211_HW_AMPDU_AGGREGATION |
IEEE80211_HW_REPORTS_TX_ACK_STATUS;
So your only option with that hardware would be to use software
encryption, and this is what the nohwcrypt module parameter will do.
If you look in the list archives I believe Yeoh had posted some
results on the performance implications of software encryption.
Cheers,
Javier
On Fri, Apr 13, 2012 at 9:01 AM, Yeoh Chun-Yeow <[email protected]> wrote:
> How about loading your kernel module rt2800usb with nohwcrypt=1.
>
> Chun-Yeow
>
> On Fri, Apr 13, 2012 at 11:00 PM, Jason Farah <[email protected]> wrote:
>> Hello all,
>>
>>
>>
>> I’m having a problem trying to get my secure mesh points talking to each
>> other. I’ve compiled authsae, I’m using linux kernel version 3.2.13 with
>> the necessary configs, and the adapters I’m working with use the rt2800usb
>> modules.
>>
>>
>>
>> In open mesh mode, everything works fine. But, I can’t seem to figure out
>> the secure mesh.
>>
>>
>>
>> First, I start up meshd-nl80211 as per the webpage. Everything seems ok
>> here except the last few lines:
>>
>>
>>
>> electrum100:~/authsae/linux# ./meshd-nl80211 -c ../config/authsae.sample.cfg
>> -s byteme -i mesh0 &
>>
>>
>>
>> ….
>>
>>
>>
>> estab with 00:14:d1:7c:33:8f
>>
>> set auth flag (seq num=1334243328)
>>
>> set plink state (seq num=1334243333)
>>
>> mesh plink with 00:14:d1:7c:33:8f established
>>
>> nlerror, cmd 11, seq 1334243330: Invalid argument
>>
>> nlerror, cmd 11, seq 1334243331: Invalid argument
>>
>> Mesh plink timer for 00:14:d1:7c:33:8f fired on state ESTAB
>>
>> Timeout for peer 00:14:d1:7c:33:8f in state 4
>>
>>
>>
>>
>>
>> It looks like an error, but appears to establish anyway? I do have the full
>> message if anyone is interested. Next, I do a station dump:
>>
>>
>>
>> electrum100:~/authsae/linux# iw dev mesh0 station dump
>>
>> Station cc:5d:4e:2b:76:d8 (on mesh0)
>>
>> inactive time: 542 ms
>>
>> rx bytes: 2148
>>
>> rx packets: 40
>>
>> tx bytes: 484
>>
>> tx packets: 3
>>
>> tx retries: 0
>>
>> tx failed: 0
>>
>> signal: -37 dBm
>>
>> signal avg: -36 dBm
>>
>> tx bitrate: 1.0 MBit/s
>>
>> mesh llid: 0
>>
>> mesh plid: 0
>>
>> mesh plink: ESTAB
>>
>> authorized: yes
>>
>> authenticated: yes
>>
>> preamble: long
>>
>> WMM/WME: yes
>>
>> MFP: yes
>>
>>
>>
>>
>>
>> The other device gives similar output. And this output looks similar to the
>> one on the o11s.org webpage. It says it’s established, but they still
>> cannot ping each other. When I go back to open mesh, everything works
>> fine. Am I missing something on the secure setup? I’m using the default
>> config file, which at first glance looks ok for me. Any pointers would be
>> greatly appreciated.
>>
>>
>>
>>
>>
>> Best regards,
>>
>> Jason Farah
>>
>>
>>
>>
>> _______________________________________________
>> Devel mailing list
>> [email protected]
>> http://lists.open80211s.org/cgi-bin/mailman/listinfo/devel
>>
> _______________________________________________
> Devel mailing list
> [email protected]
> http://lists.open80211s.org/cgi-bin/mailman/listinfo/devel
--
Javier Cardona
cozybit Inc.
http://www.cozybit.com
_______________________________________________
Devel mailing list
[email protected]
http://lists.open80211s.org/cgi-bin/mailman/listinfo/devel
