When rsb3730+nxp wifi is set to ap+station mode, ping loss and the ssid cannot be scanned.

austinlee12345

Hi NXP,

Currently setting ap+station mode on NXP wifi module + rsb3730, the following is my topology:

ap1 (ssid: 1234) -------rsb3730(station + ap (ssid:nxp_ap , ip:192.168.8.1)) --------client

The following issues have been discovered so far:

1. When the station cannot connect to ap1, the client connected to nxp_ap, will ping loss when pinging device (ping 192.168.8.1)

2. When the station cannot connect to ap1,sometimes the ssid of nxp_ap cannot be scanned for a short time, but it will be scanned after about 5 seconds.

Attached is the captured device log.

Thank you~

Austin

austinlee12345

Dear Weidong,

Yes, I have successfully set mlan0 and uap0 to ap+station. The problem I am currently encountering is: if mlan0 has been unable to connect to ap1 (ssid: 1234) , then uap0 will sometimes hang directly. If mlan0 has successfully connected to ap1 (ssid: 1234) , then everything is fine.

Attached are the steps we have taken to reproduce the problem so far.

Thank you~

BR,

Austin

weidong_sun

Dear @austinlee12345 ,

Attachment is for your reference.

The doc is very similar to your application.

Try it, please!

Thanks!

Regards,

weidong

austinlee12345

Dear Weidong,

The same problem happened today. uap0 didn't work at all, but the system worked normally. Attached is the log I captured for your reference.

Thank you~

Austin

weidong_sun

Dear @austinlee12345 ,

Did you try the steps in my document to test SAT + uap ?

Regards,

weidong

austinlee12345

Dear Weidong,

Yes, I have set ap+sta according to the document you provided. This issue will not happen every time. There is a chance that the SSID of uap0 cannot be scanned or uap0 will stop working.

Thank you~

Regards,

Austin

weidong_sun

Dear @austinlee12345 ,

I tested it for several times, It is very strange that I have not reproduced the issue you mentioned above.

-- OTA logs

Could you please provide OTA logs for me when the error occurred ?

Thanks!

Regards,

weidong

austinlee12345

Dear Weidong,

The attachment is the log I captured using dmesg in the system.

The problem of uap0 stopping may occur after repeating the steps 20~30 times.
When sets the wrong psk in wpa_supplicant config and let mlan0 cannot asscoiate to the ap, I will wait for about 5~10 minutes to observe the status of uap0.

Thank you~

Regards,

Austin

weidong_sun

Dear @austinlee12345

>.When sets the wrong psk in wpa_supplicant config and let mlan0 cannot asscoiate to the ap.

--wrong psk in wpa_supplicant is the only condition to cause the issue, right?

-- If password is correct, no issues, right?

[Additional information I need]

1. systemctl status connman.service

run the command, please! then copy logs to me.

2. If psk is not correct, connection will fail, wpa_supplicant should scan AP continuously, check it, then tell me, please!

Thanks!

Regards,

weidong

austinlee12345

Dear Weidong,

>.When sets the wrong psk in wpa_supplicant config and let mlan0 cannot asscoiate to the ap.

--wrong psk in wpa_supplicant is the only condition to cause the issue, right? As long as mlan0 cannot connect to the ap, problems may occur.

-- If password is correct, no issues, right? YES

[Additional information I need]

1. systemctl status connman.service

run the command, please! then copy logs to me. --->The attachment is the captured log

2. If psk is not correct, connection will fail, wpa_supplicant should scan AP continuously, check it, then tell me, please! --->yes,wpa_supplicant should scan AP continuously

Thank you

Regards,

Austin

weidong_sun

Dear @austinlee12345 ,

In your system, connmand ( connman.service) is also running:

====================================================

* connman.service - Connection service
Loaded: loaded (/lib/systemd/system/connman.service; enabled; vendor preset: enabled)
Active: active (running) since Thu 2024-04-18 05:33:26 UTC; 24h ago
Main PID: 641 (connmand)
Tasks: 1
Memory: 4.9M
CGroup: /system.slice/connman.service
`-641 /usr/sbin/connmand -n

====================================================

So when connection failed, scan will be continued.

Do you really need to use connmand Daemon process?

Regards,

weidong

austinlee12345

Dear Weidong

Do you mean that as long as mlan0 continues to scan, uap0 will have a chance to crash, even if they are different interfaces?

Is this a known issue with nxp wifi driver?

Regards,

Austin

weidong_sun

Dear @austinlee12345 ,

Please create the main.conf file in /etc/connman/, if there is no connman directory, create it, please!

[General]
NetworkInterfaceBlacklist=wfd0,uap0,p2p0,muap0,mwfd0,p2p1

Then restart your board. and continue your settings and operations.

Regards,

weidong

austinlee12345

Dear Weidong,

We currently use this method for verification, but we find that the problem still occurs.

Are there any other suggested methods?

Thank you~

Regards,

Austin