2020-01-29 09:22:02

by Jupiter

[permalink] [raw]
Subject: Strange inconsistant WiFi network behaviour

Hi,

I have 2 iMX6 devices running uBlox Lily WiFi, mrvl firmware
sd8801_uapsta.bin in kernel 4.19.75. In my office, a WiFi router is
just 1 meter away, the WiFi signal should not be a problem. One device
could connect to the WiFi router well and stably, one could not, here
were error messages:

[ 408.990029] ieee80211 phy0: mwifiex_cfg80211_sched_scan_start :
Invalid Sched_scan parameters
.....................

[56.986414] ieee80211 phy0: mwifiex_cfg80211_sched_scan_start :
Invalid Sched_scan parameter

connmand[13469]:
../connman-1.35/src/service.c:preferred_tech_add_by_type() type 5
service 0x119120 Telstra
connmand[13469]: ../connman-1.35/src/service.c:auto_connect_service()
preferred 1 sessions 0 reason auto
connmand[13469]: ../connman-1.35/src/ntp.c:send_timeout() send timeout
2 (retries 0)
connmand[13469]:
../connman-1.35/plugins/ofono.c:netreg_update_strength() /ubloxqmi_2
Strength 80
connmand[13469]: ../connman-1.35/src/ntp.c:send_timeout() send timeout
4 (retries 1)
connmand[13469]: ../connman-1.35/plugins/wifi.c:throw_wifi_scan()
device 0x118da8 0x116c78
connmand[13469]:
../connman-1.35/src/device.c:connman_device_ref_debug() 0x118da8 ref 4
by ../connman-1.35/plugins/wifi.c:1214:throw_wifi_scan()
connmand[13469]:
../connman-1.35/src/device.c:connman_device_set_scanning() device
0x118da8 scanning 1
connmand[13469]:
../connman-1.35/src/technology.c:__connman_technology_scan_started()
device 0x118da8
connmand[13469]: ../connman-1.35/plugins/wifi.c:autoscan_timeout() interval 27
connmand[13469]: ../connman-1.35/plugins/wifi.c:interface_state() wifi
0x11a7c8 interface state 4
connmand[13469]: ../connman-1.35/plugins/wifi.c:scan_started()
connmand[13469]: ../connman-1.35/src/rtnl.c:rtnl_message() NEWLINK len
56 type 16 flags 0x0000 seq 0 pid 0
connmand[13469]: ../connman-1.35/plugins/wifi.c:scan_finished()
connmand[13469]: ../connman-1.35/plugins/wifi.c:interface_state() wifi
0x11a7c8 interface state 3

Both devices were installed the same firmware and software, if it was
firmware / software issues, both would not be able to connect to WiFi,
as one device could connect to the office WiFi router well, it could
not the WiFi router problem either.

If I move the faulty one to my home, it could connect to my home WiFi,
so the device does not have WiFi problem, that is really confusing,
what could cause that kind problem and how to debug and fix it?

Thank you.

Kind regards,

- jh


2020-01-29 10:10:23

by Emil Petersky

[permalink] [raw]
Subject: Re: Strange inconsistant WiFi network behaviour

Try to increase distance from router or other WiFi device.

Best regards,

Emil

On 29.01.2020 10:21, JH wrote:
> Hi,
>
> I have 2 iMX6 devices running uBlox Lily WiFi, mrvl firmware
> sd8801_uapsta.bin in kernel 4.19.75. In my office, a WiFi router is
> just 1 meter away, the WiFi signal should not be a problem. One device
> could connect to the WiFi router well and stably, one could not, here
> were error messages:
>
> [ 408.990029] ieee80211 phy0: mwifiex_cfg80211_sched_scan_start :
> Invalid Sched_scan parameters
> .....................
>
> [56.986414] ieee80211 phy0: mwifiex_cfg80211_sched_scan_start :
> Invalid Sched_scan parameter
>
> connmand[13469]:
> ../connman-1.35/src/service.c:preferred_tech_add_by_type() type 5
> service 0x119120 Telstra
> connmand[13469]: ../connman-1.35/src/service.c:auto_connect_service()
> preferred 1 sessions 0 reason auto
> connmand[13469]: ../connman-1.35/src/ntp.c:send_timeout() send timeout
> 2 (retries 0)
> connmand[13469]:
> ../connman-1.35/plugins/ofono.c:netreg_update_strength() /ubloxqmi_2
> Strength 80
> connmand[13469]: ../connman-1.35/src/ntp.c:send_timeout() send timeout
> 4 (retries 1)
> connmand[13469]: ../connman-1.35/plugins/wifi.c:throw_wifi_scan()
> device 0x118da8 0x116c78
> connmand[13469]:
> ../connman-1.35/src/device.c:connman_device_ref_debug() 0x118da8 ref 4
> by ../connman-1.35/plugins/wifi.c:1214:throw_wifi_scan()
> connmand[13469]:
> ../connman-1.35/src/device.c:connman_device_set_scanning() device
> 0x118da8 scanning 1
> connmand[13469]:
> ../connman-1.35/src/technology.c:__connman_technology_scan_started()
> device 0x118da8
> connmand[13469]: ../connman-1.35/plugins/wifi.c:autoscan_timeout() interval 27
> connmand[13469]: ../connman-1.35/plugins/wifi.c:interface_state() wifi
> 0x11a7c8 interface state 4
> connmand[13469]: ../connman-1.35/plugins/wifi.c:scan_started()
> connmand[13469]: ../connman-1.35/src/rtnl.c:rtnl_message() NEWLINK len
> 56 type 16 flags 0x0000 seq 0 pid 0
> connmand[13469]: ../connman-1.35/plugins/wifi.c:scan_finished()
> connmand[13469]: ../connman-1.35/plugins/wifi.c:interface_state() wifi
> 0x11a7c8 interface state 3
>
> Both devices were installed the same firmware and software, if it was
> firmware / software issues, both would not be able to connect to WiFi,
> as one device could connect to the office WiFi router well, it could
> not the WiFi router problem either.
>
> If I move the faulty one to my home, it could connect to my home WiFi,
> so the device does not have WiFi problem, that is really confusing,
> what could cause that kind problem and how to debug and fix it?
>
> Thank you.
>
> Kind regards,
>
> - jh
>

2020-01-30 00:04:40

by Jupiter

[permalink] [raw]
Subject: Re: Strange inconsistant WiFi network behaviour

Thanks Emil and Jan,

On 1/29/20, Ryll, Jan (GED-SDD2) <[email protected]> wrote:
> Try to increase distance from router or other WiFi device.

Tried, but no avail.

> there are WiFi routers like AVM which are implement the WiFi Stack in a
> stricter way than other routers or other WiFi chips-firmware.
> I mention this cause weh ad some issue with a cypress wifi firmware which
> leads to "sometimes" disconnect from routers. And we figured out that this
> was mostly the case with AVM. We are in contact with AVM in it turns out
> that the AVM router work proberly. The issues was with the cypress chipset
> firmware. Now we are in contact with cypress and they accepted the issue.
> So it is not always a connman problem :-) .

That really worries me, my WiFi router is TP-Link AC1200 dual band
router, it should have no problem for my device to connect it as I
mentioned I have two devices one could connected without any issues.
that definitely won't be my WiFi chip uBlox Lily issues, if it was,
the another unit would not be possible to connect to my WiFi router in
office.

I am not saying it is connman problem or mwifiex problem or kernel
problem, what I like is to get helps and clues from open source
communities to help me to debug and to find issues, it could be my
contributions as well if there could be potential open source bugs to
be found from my test, debug and report :-).

Thank you.

Kind regards,

- jh
>
> -----Original Message-----
> From: JH <[email protected]>
> Sent: Wednesday, January 29, 2020 10:22 AM
> To: linux-wireless <[email protected]>; connman
> <[email protected]>
> Subject: Strange inconsistant WiFi network behaviour
>
> Hi,
>
> I have 2 iMX6 devices running uBlox Lily WiFi, mrvl firmware
> sd8801_uapsta.bin in kernel 4.19.75. In my office, a WiFi router is just 1
> meter away, the WiFi signal should not be a problem. One device could
> connect to the WiFi router well and stably, one could not, here were error
> messages:
>
> [ 408.990029] ieee80211 phy0: mwifiex_cfg80211_sched_scan_start :
> Invalid Sched_scan parameters
> .....................
>
> [56.986414] ieee80211 phy0: mwifiex_cfg80211_sched_scan_start :
> Invalid Sched_scan parameter
>
> connmand[13469]:
> ../connman-1.35/src/service.c:preferred_tech_add_by_type() type 5 service
> 0x119120 Telstra
> connmand[13469]: ../connman-1.35/src/service.c:auto_connect_service()
> preferred 1 sessions 0 reason auto
> connmand[13469]: ../connman-1.35/src/ntp.c:send_timeout() send timeout
> 2 (retries 0)
> connmand[13469]:
> ../connman-1.35/plugins/ofono.c:netreg_update_strength() /ubloxqmi_2
> Strength 80
> connmand[13469]: ../connman-1.35/src/ntp.c:send_timeout() send timeout
> 4 (retries 1)
> connmand[13469]: ../connman-1.35/plugins/wifi.c:throw_wifi_scan()
> device 0x118da8 0x116c78
> connmand[13469]:
> ../connman-1.35/src/device.c:connman_device_ref_debug() 0x118da8 ref 4 by
> ../connman-1.35/plugins/wifi.c:1214:throw_wifi_scan()
> connmand[13469]:
> ../connman-1.35/src/device.c:connman_device_set_scanning() device
> 0x118da8 scanning 1
> connmand[13469]:
> ../connman-1.35/src/technology.c:__connman_technology_scan_started()
> device 0x118da8
> connmand[13469]: ../connman-1.35/plugins/wifi.c:autoscan_timeout() interval
> 27
> connmand[13469]: ../connman-1.35/plugins/wifi.c:interface_state() wifi
> 0x11a7c8 interface state 4
> connmand[13469]: ../connman-1.35/plugins/wifi.c:scan_started()
> connmand[13469]: ../connman-1.35/src/rtnl.c:rtnl_message() NEWLINK len
> 56 type 16 flags 0x0000 seq 0 pid 0
> connmand[13469]: ../connman-1.35/plugins/wifi.c:scan_finished()
> connmand[13469]: ../connman-1.35/plugins/wifi.c:interface_state() wifi
> 0x11a7c8 interface state 3
>
> Both devices were installed the same firmware and software, if it was
> firmware / software issues, both would not be able to connect to WiFi, as
> one device could connect to the office WiFi router well, it could not the
> WiFi router problem either.
>
> If I move the faulty one to my home, it could connect to my home WiFi, so
> the device does not have WiFi problem, that is really confusing, what could
> cause that kind problem and how to debug and fix it?
>
> Thank you.
>
> Kind regards,
>
> - jh
> _______________________________________________
> connman mailing list -- [email protected] To unsubscribe send an email to
> [email protected]
>

2020-01-30 01:35:46

by Jupiter

[permalink] [raw]
Subject: Re: Strange inconsistant WiFi network behaviour

On 1/30/20, JH <[email protected]> wrote:
> Thanks Emil and Jan,
>
> On 1/29/20, Ryll, Jan (GED-SDD2) <[email protected]> wrote:
>> Try to increase distance from router or other WiFi device.
>
> Tried, but no avail.

Sorry Emil, I had spoken too soon, it is connected now. Why that
means, RF noise or interference from another device?

>> there are WiFi routers like AVM which are implement the WiFi Stack in a
>> stricter way than other routers or other WiFi chips-firmware.
>> I mention this cause weh ad some issue with a cypress wifi firmware which
>> leads to "sometimes" disconnect from routers. And we figured out that
>> this
>> was mostly the case with AVM. We are in contact with AVM in it turns out
>> that the AVM router work proberly. The issues was with the cypress
>> chipset
>> firmware. Now we are in contact with cypress and they accepted the issue.
>> So it is not always a connman problem :-) .
>
> That really worries me, my WiFi router is TP-Link AC1200 dual band
> router, it should have no problem for my device to connect it as I
> mentioned I have two devices one could connected without any issues.
> that definitely won't be my WiFi chip uBlox Lily issues, if it was,
> the another unit would not be possible to connect to my WiFi router in
> office.
>
> I am not saying it is connman problem or mwifiex problem or kernel
> problem, what I like is to get helps and clues from open source
> communities to help me to debug and to find issues, it could be my
> contributions as well if there could be potential open source bugs to
> be found from my test, debug and report :-).
>
> Thank you.
>
> Kind regards,
>
> - jh
>>
>> -----Original Message-----
>> From: JH <[email protected]>
>> Sent: Wednesday, January 29, 2020 10:22 AM
>> To: linux-wireless <[email protected]>; connman
>> <[email protected]>
>> Subject: Strange inconsistant WiFi network behaviour
>>
>> Hi,
>>
>> I have 2 iMX6 devices running uBlox Lily WiFi, mrvl firmware
>> sd8801_uapsta.bin in kernel 4.19.75. In my office, a WiFi router is just
>> 1
>> meter away, the WiFi signal should not be a problem. One device could
>> connect to the WiFi router well and stably, one could not, here were
>> error
>> messages:
>>
>> [ 408.990029] ieee80211 phy0: mwifiex_cfg80211_sched_scan_start :
>> Invalid Sched_scan parameters
>> .....................
>>
>> [56.986414] ieee80211 phy0: mwifiex_cfg80211_sched_scan_start :
>> Invalid Sched_scan parameter
>>
>> connmand[13469]:
>> ../connman-1.35/src/service.c:preferred_tech_add_by_type() type 5 service
>> 0x119120 Telstra
>> connmand[13469]: ../connman-1.35/src/service.c:auto_connect_service()
>> preferred 1 sessions 0 reason auto
>> connmand[13469]: ../connman-1.35/src/ntp.c:send_timeout() send timeout
>> 2 (retries 0)
>> connmand[13469]:
>> ../connman-1.35/plugins/ofono.c:netreg_update_strength() /ubloxqmi_2
>> Strength 80
>> connmand[13469]: ../connman-1.35/src/ntp.c:send_timeout() send timeout
>> 4 (retries 1)
>> connmand[13469]: ../connman-1.35/plugins/wifi.c:throw_wifi_scan()
>> device 0x118da8 0x116c78
>> connmand[13469]:
>> ../connman-1.35/src/device.c:connman_device_ref_debug() 0x118da8 ref 4 by
>> ../connman-1.35/plugins/wifi.c:1214:throw_wifi_scan()
>> connmand[13469]:
>> ../connman-1.35/src/device.c:connman_device_set_scanning() device
>> 0x118da8 scanning 1
>> connmand[13469]:
>> ../connman-1.35/src/technology.c:__connman_technology_scan_started()
>> device 0x118da8
>> connmand[13469]: ../connman-1.35/plugins/wifi.c:autoscan_timeout()
>> interval
>> 27
>> connmand[13469]: ../connman-1.35/plugins/wifi.c:interface_state() wifi
>> 0x11a7c8 interface state 4
>> connmand[13469]: ../connman-1.35/plugins/wifi.c:scan_started()
>> connmand[13469]: ../connman-1.35/src/rtnl.c:rtnl_message() NEWLINK len
>> 56 type 16 flags 0x0000 seq 0 pid 0
>> connmand[13469]: ../connman-1.35/plugins/wifi.c:scan_finished()
>> connmand[13469]: ../connman-1.35/plugins/wifi.c:interface_state() wifi
>> 0x11a7c8 interface state 3
>>
>> Both devices were installed the same firmware and software, if it was
>> firmware / software issues, both would not be able to connect to WiFi, as
>> one device could connect to the office WiFi router well, it could not the
>> WiFi router problem either.
>>
>> If I move the faulty one to my home, it could connect to my home WiFi, so
>> the device does not have WiFi problem, that is really confusing, what
>> could
>> cause that kind problem and how to debug and fix it?
>>
>> Thank you.
>>
>> Kind regards,
>>
>> - jh
>> _______________________________________________
>> connman mailing list -- [email protected] To unsubscribe send an email
>> to
>> [email protected]
>>
>

2020-01-30 08:29:19

by Emil Petersky

[permalink] [raw]
Subject: Re: Strange inconsistant WiFi network behaviour

On 30/01/2020 02:33, JH wrote:
> On 1/30/20, JH <[email protected]> wrote:
>> Thanks Emil and Jan,
>>
>> On 1/29/20, Ryll, Jan (GED-SDD2) <[email protected]> wrote:
>>> Try to increase distance from router or other WiFi device.
>> Tried, but no avail.
> Sorry Emil, I had spoken too soon, it is connected now. Why that
> means, RF noise or interference from another device?

Hi, yes, maybe the RF input stage is saturated by very strong signal coming
from AP / 2nd device. I have seen this on many devices/APs, when they
are in (sub-) meter distance. Exact distance when devices starts to work
properly may vary with spread / tolerances of used RF parts...

And of course, some RF input stages are less sensitive (better designed)
as the others (across various APs / devices)

Best regards,

Emil

>>> there are WiFi routers like AVM which are implement the WiFi Stack in a
>>> stricter way than other routers or other WiFi chips-firmware.
>>> I mention this cause weh ad some issue with a cypress wifi firmware which
>>> leads to "sometimes" disconnect from routers. And we figured out that
>>> this
>>> was mostly the case with AVM. We are in contact with AVM in it turns out
>>> that the AVM router work proberly. The issues was with the cypress
>>> chipset
>>> firmware. Now we are in contact with cypress and they accepted the issue.
>>> So it is not always a connman problem :-) .
>> That really worries me, my WiFi router is TP-Link AC1200 dual band
>> router, it should have no problem for my device to connect it as I
>> mentioned I have two devices one could connected without any issues.
>> that definitely won't be my WiFi chip uBlox Lily issues, if it was,
>> the another unit would not be possible to connect to my WiFi router in
>> office.
>>
>> I am not saying it is connman problem or mwifiex problem or kernel
>> problem, what I like is to get helps and clues from open source
>> communities to help me to debug and to find issues, it could be my
>> contributions as well if there could be potential open source bugs to
>> be found from my test, debug and report :-).
>>
>> Thank you.
>>
>> Kind regards,
>>
>> - jh
>>> -----Original Message-----
>>> From: JH <[email protected]>
>>> Sent: Wednesday, January 29, 2020 10:22 AM
>>> To: linux-wireless <[email protected]>; connman
>>> <[email protected]>
>>> Subject: Strange inconsistant WiFi network behaviour
>>>
>>> Hi,
>>>
>>> I have 2 iMX6 devices running uBlox Lily WiFi, mrvl firmware
>>> sd8801_uapsta.bin in kernel 4.19.75. In my office, a WiFi router is just
>>> 1
>>> meter away, the WiFi signal should not be a problem. One device could
>>> connect to the WiFi router well and stably, one could not, here were
>>> error
>>> messages:
>>>
>>> [ 408.990029] ieee80211 phy0: mwifiex_cfg80211_sched_scan_start :
>>> Invalid Sched_scan parameters
>>> .....................
>>>
>>> [56.986414] ieee80211 phy0: mwifiex_cfg80211_sched_scan_start :
>>> Invalid Sched_scan parameter
>>>
>>> connmand[13469]:
>>> ../connman-1.35/src/service.c:preferred_tech_add_by_type() type 5 service
>>> 0x119120 Telstra
>>> connmand[13469]: ../connman-1.35/src/service.c:auto_connect_service()
>>> preferred 1 sessions 0 reason auto
>>> connmand[13469]: ../connman-1.35/src/ntp.c:send_timeout() send timeout
>>> 2 (retries 0)
>>> connmand[13469]:
>>> ../connman-1.35/plugins/ofono.c:netreg_update_strength() /ubloxqmi_2
>>> Strength 80
>>> connmand[13469]: ../connman-1.35/src/ntp.c:send_timeout() send timeout
>>> 4 (retries 1)
>>> connmand[13469]: ../connman-1.35/plugins/wifi.c:throw_wifi_scan()
>>> device 0x118da8 0x116c78
>>> connmand[13469]:
>>> ../connman-1.35/src/device.c:connman_device_ref_debug() 0x118da8 ref 4 by
>>> ../connman-1.35/plugins/wifi.c:1214:throw_wifi_scan()
>>> connmand[13469]:
>>> ../connman-1.35/src/device.c:connman_device_set_scanning() device
>>> 0x118da8 scanning 1
>>> connmand[13469]:
>>> ../connman-1.35/src/technology.c:__connman_technology_scan_started()
>>> device 0x118da8
>>> connmand[13469]: ../connman-1.35/plugins/wifi.c:autoscan_timeout()
>>> interval
>>> 27
>>> connmand[13469]: ../connman-1.35/plugins/wifi.c:interface_state() wifi
>>> 0x11a7c8 interface state 4
>>> connmand[13469]: ../connman-1.35/plugins/wifi.c:scan_started()
>>> connmand[13469]: ../connman-1.35/src/rtnl.c:rtnl_message() NEWLINK len
>>> 56 type 16 flags 0x0000 seq 0 pid 0
>>> connmand[13469]: ../connman-1.35/plugins/wifi.c:scan_finished()
>>> connmand[13469]: ../connman-1.35/plugins/wifi.c:interface_state() wifi
>>> 0x11a7c8 interface state 3
>>>
>>> Both devices were installed the same firmware and software, if it was
>>> firmware / software issues, both would not be able to connect to WiFi, as
>>> one device could connect to the office WiFi router well, it could not the
>>> WiFi router problem either.
>>>
>>> If I move the faulty one to my home, it could connect to my home WiFi, so
>>> the device does not have WiFi problem, that is really confusing, what
>>> could
>>> cause that kind problem and how to debug and fix it?
>>>
>>> Thank you.
>>>
>>> Kind regards,
>>>
>>> - jh
>>> _______________________________________________
>>> connman mailing list -- [email protected] To unsubscribe send an email
>>> to
>>> [email protected]
>>>


--
Emil Petersky

________________________________________________________________________

StreamUnlimited Engineering GmbH
High Tech Campus Vienna, Gutheil-Schoder-Gasse 10, 1100 Vienna, Austria
Office: +43 1 667 2002 4679 Fax: +43 1 667 2002 4401
Mail to: [email protected]
Visit us: http://www.streamunlimited.com

Meet us at:
ISE - Amsterdam, 11 - 14 February, booth 7-X195
The Bristol Show - Bristol, 21 - 23 February
Embedded World - Nuremberg, 25 - 27 February, booth 3-501
Light & Building - Frankfurt, 8 - 13 March