Skip to contentSkip to conversation

Client station sends probes on DFS channels

DISCUSSION Jean-Pierre Tosoni 2 participants 4 messages Nov 30, 2016 – Dec 5, 2016 7 messages not in the archive Qualcomm Atheros ATH10K wireless driver

lore
Share & tools
lore
#0 lore raw

Hello list,

There is a case where I can see probes on a DFS channel, from a non-associated station using ath10k. (note that the problem does not arise when using ath9k).

*The setup*

I am using Openwrt, wpa_supplicant and compat-wireless 2016-10-08,
Card firmware is ath10k_pci: qca988x hw2.0 (0x4100016c, 0x043202ff)
fw 10.2.4.70-2 api 5 htt-ver 2.1 wmi-op 5 htt-op 2 cal otp max-sta 128 features no-p2p
ath10k_pci: debug 0 debugfs 1 tracing 0 dfs 1 testmode 1

I am using channel 116, regdom US or FR, where I see no traffic at all using wireshark+Airpcap.
I set wpa_supplicant to scan this channel only for a specific SSID "ssid1".

At initial startup of the client device, no probes are going out, which is OK.

Then, on another device, I start a hostapd set to channel 116, with a different SSID "otherssid" so that the supplicant will not associate.

Shortly (1-2s) after the beacons appear in the air, the client begins to Send probe request in the air, which is unexpected, but acceptable since the client can infer absence of radars from the presence of beacons.

*The problem*

If I power down the AP, the client continues to send probes for around 10 minutes, which is unacceptable since it cannot handle radar detection, as it is a slave device in the meaning of ETSI/EN 301-893[1].

Some tests I made:
- I tried to investigate the "beacon hint" mechanism but it appears that it is not used on DFS channels.

- I tried to force the IEEE80211_NO_IR flag when IEEE80211_CHAN_DFS is set.

- When I reload the reg. domain using "iw reg set", the probes cease, but will
reappear if I cycle my AP again On then Off.

- When I let the client associate, then disassociate and stop the AP, the same
problem arises. It disappears if I add a call to ath10k_regd_update() in
mac.c
after a disconnection (This is not a fix, since in my case the client
never
associates).

- Since at startup, the client does not send probes, I infer that the problem
is *not* caused by a hidden AP that the card could see but not airpcap.

- I tried with channels 52 and 100, with regdom FR or US: same problem.

Any ideas?

[1]
http://www.etsi.org/deliver/etsi_en/301800_301899/301893/01.08.01_60/en_3018
93v010801p.pdf

J.P. Tosoni - ACKSYS

_______________________________________________
ath10k mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/ath10k

A message that is not in the archive (sent off-list or not yet received): CAJ-Vmom1EDyHH8Pcuxq83ptsYEGj0LWeDCNvx7gDAUKQmbWZ6g@mail.gmail.com

#2 lore raw

Hi Adrian,

Quoted 1 line

What's the etsi spec say about once a station hears beacons?

What ETSI EN301893 v1.8.1 [1] says is:

1) ETSI distinguishes "master" devices and "slave" devices (section 4.7.1.3)

2) "Slaves shall only operate in a network controlled by (...) a master" (section 4.7.1.3)

3) "A slave device shall stop its transmissions on a channel whenever instructed by a master device" (section 4.7.1.4)
=> In my understanding, from a 802.11 point of view, this implies that
The slave obeys the master, hence it is associated with it "somehow". Otherwise, the station must be considered as being a master.

4) Channel move time: the slave must stop transmission within *10 seconds* After being instructed so (section 4.7.2.5.1 and table D.1)

5) During these 10 seconds, the airtime used must not exceed 1 second.

So,
- either any unrelated AP is considered a master, and when beacons cease this should be considered a hint to stop within 10 seconds,
- or the unassociated client is considered a master itself, and must obey CAC, NOP and so on.

Hence I think the relevant duration is the "channel move time" in table D.1 which is limited to 10 seconds -- instead of 10 minutes.

Jean-Pierre

De : Adrian Chadd [mailto:[email protected]] Envoyé : mercredi 30 novembre 2016 21:41
À : Jean-Pierre Tosoni
Cc : [email protected]
Objet : Re: Client station sends probes on DFS channels

Hi,
I thought that seeing beacons meant the sea could treat it as non passive for a while. Ten minutes sounds about right. What's the etsi spec say about once a station hears beacons?
A

On Nov 30, 2016 12:36 PM, "Jean-Pierre Tosoni" <[email protected]> wrote:
Hello list,

There is a case where I can see probes on a DFS channel, from a non-associated station using ath10k. (note that the problem does not arise when using ath9k).

*The setup*

I am using Openwrt, wpa_supplicant and compat-wireless 2016-10-08, Card firmware is ath10k_pci: qca988x hw2.0 (0x4100016c, 0x043202ff)
fw 10.2.4.70-2 api 5 htt-ver 2.1 wmi-op 5 htt-op 2 cal otp max-sta 128 features no-p2p
ath10k_pci: debug 0 debugfs 1 tracing 0 dfs 1 testmode 1

I am using channel 116, regdom US or FR, where I see no traffic at all using wireshark+Airpcap.
I set wpa_supplicant to scan this channel only for a specific SSID "ssid1".

At initial startup of the client device, no probes are going out, which is OK.

Then, on another device, I start a hostapd set to channel 116, with a different SSID "otherssid" so that the supplicant will not associate.

Shortly (1-2s) after the beacons appear in the air, the client begins to Send probe request in the air, which is unexpected, but acceptable since the client can infer absence of radars from the presence of beacons.

*The problem*

If I power down the AP, the client continues to send probes for around 10 minutes, which is unacceptable since it cannot handle radar detection, as it is a slave device in the meaning of ETSI/EN 301-893[1].

Some tests I made:
- I tried to investigate the "beacon hint" mechanism but it appears that it is not used on DFS channels.

- I tried to force the IEEE80211_NO_IR flag when IEEE80211_CHAN_DFS is set.

- When I reload the reg. domain using "iw reg set", the probes cease, but will reappear if I cycle my AP again On then Off.

- When I let the client associate, then disassociate and stop the AP, the same problem arises. It disappears if I add a call to ath10k_regd_update() in mac.c after a disconnection (This is not a fix, since in my case the client never associates).

- Since at startup, the client does not send probes, I infer that the problem is *not* caused by a hidden AP that the card could see but not airpcap.

- I tried with channels 52 and 100, with regdom FR or US: same problem.

Any ideas?

[1]
http://www.etsi.org/deliver/etsi_en/301800_301899/301893/01.08.01_60/en_3018
93v010801p.pdf

J.P. Tosoni - ACKSYS

_______________________________________________
ath10k mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/ath10k

_______________________________________________
ath10k mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/ath10k

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): CAJ-VmokpLMHL3ND34C4FAW3tVtxHKKOcgSHR9Vs+GsxPwMVE8A@mail.gmail.com

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): CAJ-Vmo=DabqVwVpBui7gu9p9EBQ_o+o7JGSaKriHvA-OMWBb0A@mail.gmail.com

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): CAJ-VmonyohH8qEGMYT5fVAgQ9BskZYj9djYU6mZ1Zp_HEsgdVw@mail.gmail.com

#10 lore raw

Just keeping the list informed... At some point the discussion turned private.

Jean-Pierre Tosoni

-----Message d'origine-----
De : Adrian Chadd [mailto:[email protected]] Envoyé : jeudi 1 décembre 2016 19:20
À : Jean-Pierre Tosoni
Objet : Re: Client station sends probes on DFS channels

well, if it fails ETSI certification then you can file a complain with QCA. :) they're passing certification tests with it atm, so...

-a

Quoted 90 lines · Jean-Pierre Tosoni

:(
Can we expect a fix someday?

JP

Quoted 82 lines

-----Message d'origine-----
De : Adrian Chadd [mailto:[email protected]] Envoyé : jeudi 1 décembre 2016 19:00 À : Jean-Pierre Tosoni Objet : Re: Client station sends probes on DFS channels

firmware

-a

Quoted 69 lines · Jean-Pierre Tosoni

Hi again,

That being said, is this 10-minutes delay implemented in ath10k or in the firmware of the card? Or is it configurable maybe?

Jean-Pierre

Quoted 61 lines

-----Message d'origine-----
De : Adrian Chadd [mailto:[email protected]] Envoyé : jeudi 1 décembre 2016 18:21 À : Jean-Pierre Tosoni Objet : Re: Client station sends probes on DFS channels

Hi,

So the implementations I've seen all treat "I see beacons" as "the channel is blessed"; there's no logic for doing it for an AP that you're associated to.

That's likely what we need clarification around.

The "controlled by a master" doesn't necessarily imply that the STA is controlled by the AP, just that the AP has gone through the CAC and is doing DFS checking. Otherwise the STA couldn't actually send assoc req/response frames, because the station hasn't yet associated to an AP to be considered "in control" by it.

-adrian

Quoted 6 lines · Jean-Pierre Tosoni

Not quite so.

Quoted 1 line

* if you get a CSA, the STA should move and not use the air

No, because the STA is not associated: it is currently searching for an

AP.

Quoted 28 lines

So, it cannot receive a CSA, and should do a passive scan.

Quoted 2 lines

* if you just start another AP And then kill the APs, there's nothing above that says when the STA should stop talking?

Wrong, the norm says that the STA should *not begin* talking (since it is not associated with a master AP)

Quoted 19 lines

-----Message d'origine-----
De : Adrian Chadd [mailto:[email protected]] Envoyé :
jeudi 1 décembre 2016 17:53 À : Jean-Pierre Tosoni Cc :
[email protected] Objet : Re: Client station sends probes on DFS channels

Hi,

So to be clear:

* if you get a CSA, the STA should move and not use the air
* if you just start another AP And then kill the APs, there's nothing above that says when the STA should stop talking?

Thanks,

-adrian

_______________________________________________
ath10k mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/ath10k