[rt2x00-users] rt2800pci works nearly fine with 2.6.39rc7 kernel and compat-wireless-2.6.39-rc6-1 - problems during rekeying of PTK

[Helmut Schaa]

Hi,

On Wed, May 25, 2011 at 10:26 PM, Gertjan van Wingerde
<gwingerde at gmail.com> wrote:
> On 05/25/11 20:12, Andreas Hartmann wrote:
>> Gertjan van Wingerde schrieb:
>>> On 05/24/11 21:50, Andreas Hartmann wrote:
>>>> PTK rekeying didn't work throughout the whole test:
>>>>
>>>>
>>>> May 24 19:50:04 pc hostapd: wlan1: STA 48:5d:60:11:22:33 WPA: rekeying PTK
>>>> May 24 19:50:04 pc hostapd: wlan1: STA 48:5d:60:11:22:33 WPA: sending 1/4 msg of 4-Way Handshake
>>>> May 24 19:50:04 pc hostapd: wlan1: STA 48:5d:60:11:22:33 WPA: received EAPOL-Key frame (2/4 Pairwise)
>>>> May 24 19:50:04 pc hostapd: wlan1: STA 48:5d:60:11:22:33 WPA: sending 3/4 msg of 4-Way Handshake
>>>> May 24 19:50:05 pc hostapd: wlan1: STA 48:5d:60:11:22:33 WPA: EAPOL-Key timeout
>>>> May 24 19:50:05 pc hostapd: wlan1: STA 48:5d:60:11:22:33 WPA: sending 3/4 msg of 4-Way Handshake
>>>> May 24 19:50:06 pc hostapd: wlan1: STA 48:5d:60:11:22:33 WPA: EAPOL-Key timeout
>>>> May 24 19:50:06 pc hostapd: wlan1: STA 48:5d:60:11:22:33 WPA: sending 3/4 msg of 4-Way Handshake
>>>> May 24 19:50:07 pc hostapd: wlan1: STA 48:5d:60:11:22:33 WPA: EAPOL-Key timeout
>>>> May 24 19:50:07 pc hostapd: wlan1: STA 48:5d:60:11:22:33 WPA: sending 3/4 msg of 4-Way Handshake
>>>> May 24 19:50:08 pc hostapd: wlan1: STA 48:5d:60:11:22:33 WPA: EAPOL-Key timeout
>>>> May 24 19:50:08 pc hostapd: wlan1: STA 48:5d:60:11:22:33 WPA: PTKINITNEGOTIATING: Retry limit 4 reached
>>>> May 24 19:50:08 pc hostapd: wlan1: STA 48:5d:60:11:22:33 IEEE 802.11: deauthenticated due to local deauth request
>>>>
>>>> This test was done with a ath9k based supplicant (ar9285).
>>>
>>> I guess packets can always get lost, so broken rekeying can also indicate a temporary occurrence of noise on the channel. How often does
>>> the rekeying fail and how long does it take between two broken rekeyings?
>>
>> As said above: it never worked during 100 PTK-rekeyings!
>
> Ah, OK. I took your statement as saying PTK rekeying doesn't work all the time.
> Re-reading it now I see it can also indicate "doesn't work at all" ;-)

Andreas, could you please check if forcing ICMP frames to other queues while
running your netperf test on AC_BE causes ICMP frames to also get lost/delayed?

"ping host" should send them over AC_BE (same queue as your netperf data)
"ping -Q 150 host" should force them over AC_VI
"ping -Q 200 host" should use AC_VO

Thanks,
Helmut