user12345
Posts: 59
Joined: Mon Dec 22, 2014 7:06 pm

RPI3 USB Hardware Bug?

Fri Mar 02, 2018 9:10 am

I had a RPI1 before, replace it with a RPI3. Same Powersupply, same 2M USB Cable and a Alpha Wlan Stick.
RPI1 worked 24/7.

Wlan stops working after a Random Time(there are no other programs running on the Device), nothing in the logs or dmesg.
No Data can be received, the device wlan0/mon0 is still there. You can start airmon-ng on it again but it shows no
Data.

I tryed:
-Tryed different Kernel Versions
-Change the USB Cable
-Exchange the Capicator in the Powersupply (Voltage on the wlan stick is good)
-Add additional capicators in the wlan stick(controller chip have anyway a 3V3 Regulator only the Antenna Amp runs on 5V)
and add higher value resistors to the LEDs(reduce power)
-Remove the onboad Wlan driver
-All other known wifi realated bug stuff like Powersaving on/off

It works again if you restart the whole System or
modprobe -rt2800usb, sleep (2), modprobe rt2800usb

Its seems that with lower temprature the problem happends more often. Sometimes it happens after 2-3 Month. But i had it about all 2 days now.
A webcam on the rpi3 works fine 24/7 about 1.5M Cable.
Is it possible that the USB Controller/Hub dont follow the Specs and is unable to drive a 2M USB Cable?

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 22695
Joined: Sat Jul 30, 2011 7:41 pm

Re: RPI3 USB Hardware Bug?

Fri Mar 02, 2018 10:32 am

The USB on the Pi3 is the same as that on the Pi1 IIRC. I would not expect there to be differences in the HW causing issues.

More likely to be an issue with the drivers - did you update the OS at the same time as moving to a Pi3?

Are there any error messages in syslog?
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Contrary to popular belief, humorous signatures are allowed. Here's an example...
"My grief counseller just died, luckily, he was so good, I didn't care."

User avatar
HawaiianPi
Posts: 4208
Joined: Mon Apr 08, 2013 4:53 am
Location: Aloha, Oregon USA

Re: RPI3 USB Hardware Bug?

Fri Mar 02, 2018 11:40 am

What are the spec's of the PSU.

The Pi3B requires significantly more power than the old B+, so a PSU that worked fine on the B+ is not necessarily okay for use on a Pi3B. Combine a PSU that can't maintain 5V+ under load with long USB cables and you'll get exactly what you are observing. The micro-USB cable supplying the power to the Pi3B could also be an issue if it's a typical USB 2.0 cable with a wire gauge of 28 or higher (higher gauge number = thinner wires).
My mind is like a browser. 27 tabs are open, 9 aren't responding,
lots of pop-ups...and where is that annoying music coming from?

user12345
Posts: 59
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Fri Mar 02, 2018 4:16 pm

jamesh wrote:
Fri Mar 02, 2018 10:32 am
The USB on the Pi3 is the same as that on the Pi1 IIRC. I would not expect there to be differences in the HW causing issues.

More likely to be an issue with the drivers - did you update the OS at the same time as moving to a Pi3?

Are there any error messages in syslog?
That cant be true. If you attach a Television Receiver to RPI1 it fails(tested with 3 PSU). There are stripes in the Picture, more if you use LAN. Means there is another USB Bug. And sometimes it drops the Webcam. I read on the Internet of USB Bugs from RPI1. On RPI3 webcam works 24/7.

I had to update the OS (Rasbian) cause the RPI1 SD Card doesnt Boot with the old Raspbian Version. I tryed alot of Rasbian Versions all fail.
No no usb/wlan related, just CRON Spam. Only after restarting i get:
Mar 1 21:45:44 rpi3 kernel: [95679.314431] usbcore: deregistering interface driver rt2800usb
Mar 1 21:45:44 rpi3 kernel: [95679.606858] ieee80211 phy0: rt2x00queue_flush_queue: Warning - Queue 14 failed to flush
Mar 1 21:45:47 rpi3 kernel: [95682.506854] usb 1-1.4: reset high-speed USB device number 5 using dwc_otg
Mar 1 21:45:47 rpi3 kernel: [95682.647290] ieee80211 phy0: rt2x00_set_rt: Info - RT chipset 3070, rev 0201 detected
Mar 1 21:45:47 rpi3 kernel: [95682.668145] ieee80211 phy0: rt2x00_set_rf: Info - RF chipset 0005 detected
Mar 1 21:45:47 rpi3 kernel: [95682.668845] ieee80211 phy0: Selected rate control algorithm 'minstrel_ht'
Mar 1 21:45:47 rpi3 kernel: [95682.670332] usbcore: registered new interface driver rt2800usb
Mar 1 21:45:50 rpi3 kernel: [95686.070976] ieee80211 phy0: rt2x00lib_request_firmware: Info - Loading firmware file 'rt2870.bin'
Mar 1 21:45:50 rpi3 kernel: [95686.071232] ieee80211 phy0: rt2x00lib_request_firmware: Info - Firmware detected - version: 0.29
Mar 1 21:45:59 rpi3 kernel: [95695.077834] device mon0 entered promiscuous mode

user12345
Posts: 59
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Fri Mar 02, 2018 4:46 pm

HawaiianPi wrote:
Fri Mar 02, 2018 11:40 am
What are the spec's of the PSU.
Its one that are supplied for RPI about 1,5A. It had a 680µF/10V capicator, i replaced it with 1000µF/10V cause of this Bug.
The Power is in the Spec 4.88V on the Wifi Stick(after the 2M Cable). It uses 5V only for Antenna Amplifier, the controller has a 3V3 Regulator means it would even not crash at about 4.50V. 1 Reason to buy a RPI3 is the less Power consumption(Step-Down instead of Linear). The RPI1 was 3.8W Idle, RPI3 is 3.2W All common Tasks running. Its set to 900MHZ max and 550MHZ min. Even with Mouse+Keyboard+HDMI connected it doesnt go above 4.2W. Thats the Full Power Consumption on the 230V side with all ‎losses.
The PSU too cant have short voltage drops cause the attached USV has some ms startup Time i added about 100000µF Highcap buffer and the
USV has about 4700µF in the 5V Line too.

User avatar
FTrevorGowen
Forum Moderator
Forum Moderator
Posts: 4947
Joined: Mon Mar 04, 2013 6:12 pm
Location: Bristol, U.K.
Contact: Website

Re: RPI3 USB Hardware Bug?

Fri Mar 02, 2018 7:33 pm

user12345 wrote:
Fri Mar 02, 2018 4:46 pm
HawaiianPi wrote:
Fri Mar 02, 2018 11:40 am
What are the spec's of the PSU.
Its one that are supplied for RPI about 1,5A. It had a 680µF/10V capicator, i replaced it with 1000µF/10V cause of this Bug.
The Power is in the Spec 4.88V on the Wifi Stick(after the 2M Cable). It uses 5V only for Antenna Amplifier, the controller has a 3V3 Regulator means it would even not crash at about 4.50V. 1 Reason to buy a RPI3 is the less Power consumption(Step-Down instead of Linear). The RPI1 was 3.8W Idle, RPI3 is 3.2W All common Tasks running. Its set to 900MHZ max and 550MHZ min. Even with Mouse+Keyboard+HDMI connected it doesnt go above 4.2W. Thats the Full Power Consumption on the 230V side with all ‎losses.
The PSU too cant have short voltage drops cause the attached USV has some ms startup Time i added about 100000µF Highcap buffer and the
USV has about 4700µF in the 5V Line too.
Whilst a "good" 5V 1.5A PSU was OK for P1 Pi's from the B+ onwards, if any significant current (upto 300mA per port) were to be required by USB peripherals a "good" 5V 2A PSU** was recommended (together with a good quality, short, USB lead). To get best performance from a P3B the recommended current was increased to 2.5A. OOC, who supplied your 1.5A supply?
Trev.
** http://www.cpmspectrepi.uk/raspberry_pi ... _Models.29
(There are links in that page to my "current loading tests" of quite a few "power supplies" and "phone chargers" - the latter often not good for "power supply" use because their voltage drops "faster" above ~1A)
Still running Raspbian Jessie on some older Pi's (an A, B1, B2, B+, P2B, 3xP0, P0W) but Stretch on my 2xP3A+, P3B+, P3B, B+, A+ and a B2. See: https://www.cpmspectrepi.uk/raspberry_pi/raspiidx.htm

user12345
Posts: 59
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Fri Mar 02, 2018 8:10 pm

FTrevorGowen wrote:
Fri Mar 02, 2018 7:33 pm
To get best performance from a P3B the recommended current was increased to 2.5A. OOC, who supplied your 1.5A supply?
Its from reichelt.
If the RPI3 require 3.2W and the PSU can continues deliver 7W all is fine. The Voltages are checked [email protected]
>5V at RPI3. Allowed is 5V +-5%, means down to 4.75V is in spec.
That are valid voltages and cant be a problem.
But i let it run now from USV, that delivers 3A continues power(15W) and see what happends.

User avatar
FTrevorGowen
Forum Moderator
Forum Moderator
Posts: 4947
Joined: Mon Mar 04, 2013 6:12 pm
Location: Bristol, U.K.
Contact: Website

Re: RPI3 USB Hardware Bug?

Fri Mar 02, 2018 8:36 pm

user12345 wrote:
Fri Mar 02, 2018 8:10 pm
FTrevorGowen wrote:
Fri Mar 02, 2018 7:33 pm
To get best performance from a P3B the recommended current was increased to 2.5A. OOC, who supplied your 1.5A supply?
Its from reichelt.
If the RPI3 require 3.2W and the PSU can continues deliver 7W all is fine. The Voltages are checked [email protected]
>5V at RPI3. Allowed is 5V +-5%, means down to 4.75V is in spec.
That are valid voltages and cant be a problem.
But i let it run now from USV, that delivers 3A continues power(15W) and see what happends.
It depends upon how those "valid voltages" are measured. Whilst my measurements are of "static current loading" (using a DMM and resistive loads) dynamic/transient loading effects can be inferred from the, albeit extrapolated, "slope resistance". Ideally (ie. if I still had access to one) I would check the transient loading with an oscilloscope. But that's not really essential since a Pi (since the B+) has built-in hardware & software "low-voltage detection". NB: the specs. for the 5V rail do not effect the Pi itself, but do relate to the tolerance of any attached USB devices which, if they're connected by long, poor quality, USB cables will suffer from transient voltage droops (associated with transient current surges) which will not been detected by the Pi's circuitry or a DMM at "their end". One thing you haven't commented on is exactly what current those attached USB devices need (NB: current, not, power) or, as a crude guide, what is reported back through the Pi via the usb-devices command** . Remember the total current available is 1200mA shared over the four USB ports when using the recommended supply.
Trev.
** http://www.cpmspectrepi.uk/raspberry_pi ... erReq.html
Unfortunately it's not as useful as one would hope since often all that is reported are the low/100mA, medium/200mA or high/500mA values - I've only come across what might be measured/specified "values" for some USB flash drives & Card Readers:
http://www.cpmspectrepi.uk/raspberry_pi ... rives.html
http://www.cpmspectrepi.uk/raspberry_pi ... aders.html
Still running Raspbian Jessie on some older Pi's (an A, B1, B2, B+, P2B, 3xP0, P0W) but Stretch on my 2xP3A+, P3B+, P3B, B+, A+ and a B2. See: https://www.cpmspectrepi.uk/raspberry_pi/raspiidx.htm

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 22695
Joined: Sat Jul 30, 2011 7:41 pm

Re: RPI3 USB Hardware Bug?

Sat Mar 03, 2018 10:25 am

user12345 wrote:
Fri Mar 02, 2018 4:16 pm
jamesh wrote:
Fri Mar 02, 2018 10:32 am
The USB on the Pi3 is the same as that on the Pi1 IIRC. I would not expect there to be differences in the HW causing issues.

More likely to be an issue with the drivers - did you update the OS at the same time as moving to a Pi3?

Are there any error messages in syslog?
That cant be true. If you attach a Television Receiver to RPI1 it fails(tested with 3 PSU). There are stripes in the Picture, more if you use LAN. Means there is another USB Bug. And sometimes it drops the Webcam. I read on the Internet of USB Bugs from RPI1. On RPI3 webcam works 24/7.

I had to update the OS (Rasbian) cause the RPI1 SD Card doesnt Boot with the old Raspbian Version. I tryed alot of Rasbian Versions all fail.
No no usb/wlan related, just CRON Spam. Only after restarting i get:
Mar 1 21:45:44 rpi3 kernel: [95679.314431] usbcore: deregistering interface driver rt2800usb
Mar 1 21:45:44 rpi3 kernel: [95679.606858] ieee80211 phy0: rt2x00queue_flush_queue: Warning - Queue 14 failed to flush
Mar 1 21:45:47 rpi3 kernel: [95682.506854] usb 1-1.4: reset high-speed USB device number 5 using dwc_otg
Mar 1 21:45:47 rpi3 kernel: [95682.647290] ieee80211 phy0: rt2x00_set_rt: Info - RT chipset 3070, rev 0201 detected
Mar 1 21:45:47 rpi3 kernel: [95682.668145] ieee80211 phy0: rt2x00_set_rf: Info - RF chipset 0005 detected
Mar 1 21:45:47 rpi3 kernel: [95682.668845] ieee80211 phy0: Selected rate control algorithm 'minstrel_ht'
Mar 1 21:45:47 rpi3 kernel: [95682.670332] usbcore: registered new interface driver rt2800usb
Mar 1 21:45:50 rpi3 kernel: [95686.070976] ieee80211 phy0: rt2x00lib_request_firmware: Info - Loading firmware file 'rt2870.bin'
Mar 1 21:45:50 rpi3 kernel: [95686.071232] ieee80211 phy0: rt2x00lib_request_firmware: Info - Firmware detected - version: 0.29
Mar 1 21:45:59 rpi3 kernel: [95695.077834] device mon0 entered promiscuous mode
The USB controller on the Soc's across the Pi range are the same. I think the hub chip may have changed over time.
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Contrary to popular belief, humorous signatures are allowed. Here's an example...
"My grief counseller just died, luckily, he was so good, I didn't care."

user12345
Posts: 59
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Fri Mar 09, 2018 9:02 pm

FTrevorGowen wrote:
Fri Mar 02, 2018 8:36 pm
One thing you haven't commented on is exactly what current those attached USB devices need (NB: current, not, power) or, as a crude guide, what is reported back through the Pi via the usb-devices command** . Remember the total current available is 1200mA shared over the four USB ports when using the recommended supply.
500mA&450mA you know that is usual not a correct value. 950mA*5V would be 4.75W, thats more as the whole PI&USB&PSU require(3.2W).

user12345
Posts: 59
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Fri Mar 09, 2018 9:14 pm

jamesh wrote:
Sat Mar 03, 2018 10:25 am
The USB controller on the Soc's across the Pi range are the same. I think the hub chip may have changed over time.
Is USB not included in the CPU/SOC? And the CPU is different on PI1/PI3.

Okay cause the USV deliver 3A/15W and the USB device drops death again its the Hardware/SOC/PI-HUB or they build in a new Bug in all new Kernels (the old PI1 Kernel could´t boot on PI3. Cause i tryed the first that works on PI3, the newest and some between).

On another Linux Computer i seen such message:
[3505310.281823] ieee80211 phy0: rt2x00usb_watchdog_tx_dma: Warning - TX queue 0 DMA timed out, invoke forced forced reset
but never on PI. On the real computer there is no drop/signal lost or any other problems. Maybe the developers forget to inbuild a required watchdog or broke it?

User avatar
rpdom
Posts: 14424
Joined: Sun May 06, 2012 5:17 am
Location: Chelmsford, Essex, UK

Re: RPI3 USB Hardware Bug?

Fri Mar 09, 2018 10:25 pm

user12345 wrote:
Fri Mar 09, 2018 9:14 pm
jamesh wrote:
Sat Mar 03, 2018 10:25 am
The USB controller on the Soc's across the Pi range are the same. I think the hub chip may have changed over time.
Is USB not included in the CPU/SOC? And the CPU is different on PI1/PI3.
The USB is part of the SoC. It is not part of the CPU.

The SoC includes the CPU(s), and the GPU along with USB, GPIO and other circuit blocks. Other than the CPU(s), the boot ROM, and a few other fixes, nothing else has changed on the SoC.

User avatar
FTrevorGowen
Forum Moderator
Forum Moderator
Posts: 4947
Joined: Mon Mar 04, 2013 6:12 pm
Location: Bristol, U.K.
Contact: Website

Re: RPI3 USB Hardware Bug?

Sat Mar 10, 2018 5:25 pm

user12345 wrote:
Fri Mar 09, 2018 9:02 pm
FTrevorGowen wrote:
Fri Mar 02, 2018 8:36 pm
One thing you haven't commented on is exactly what current those attached USB devices need (NB: current, not, power) or, as a crude guide, what is reported back through the Pi via the usb-devices command** . Remember the total current available is 1200mA shared over the four USB ports when using the recommended supply.
500mA&450mA you know that is usual not a correct value. 950mA*5V would be 4.75W, thats more as the whole PI&USB&PSU require(3.2W).
W.r.t. "thats more as the whole PI&USB&PSU require(3.2W)" why does that "surprise you"? The current drawn from the 5V supply from the USB ports, whilst passing through the same polyfuse and other protection circuits** of the Pi's "on-board" 3.3V & 1.8V supplies (the SoC's GPIO's are 3.3V "devices") is "independent" of that "for the Pi". For a P3B a 5V 2.5A supply is recommended ie. one with a slightly higher current rating than the 2.4A of the polyfuse. 1.2A of that is available, but shared over the four USB 2.0 ports, for USB use and the remainder for the Pi, its on board peripherals and any GPIO driven modules (which, whilst requiring to use 3.3V "logic circuitry" can also use the 5V supply for "other purposes"). Since USB 2.0 specifies a "minimum" voltage of 4.75V**** that's where the 5V +/- 0.25V spec. for the PSU comes from. The low-voltage detection circuit (and it's associated desktop icon display software) monitors and warns of both long-term and transient voltage drops caused by current surges mainly through USB devices (eg. HDD motors spinning up, hot-plugging ...) . Such voltage drops can be made worse by using long, poor quality (thin, high-resistance) USB power cables. Should a transient drop be sufficient to cause writes to the uSDHC card to fail data corruption may occur (and, eventually, card/system failure). The circuitry/software in use has existed since the B+ and is thus "well-established"!
Trev.
** NB: They do not exist on a P0, P0W or P0WH
**** Some USB devices operate at 3.3V internally.
Still running Raspbian Jessie on some older Pi's (an A, B1, B2, B+, P2B, 3xP0, P0W) but Stretch on my 2xP3A+, P3B+, P3B, B+, A+ and a B2. See: https://www.cpmspectrepi.uk/raspberry_pi/raspiidx.htm

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 22695
Joined: Sat Jul 30, 2011 7:41 pm

Re: RPI3 USB Hardware Bug?

Sat Mar 10, 2018 5:58 pm

Just to fill in some gaps.

There is ONE USB port of the SoC's used on Pi0,1,2,3. On the 1,2 and 3 this is plugged in to a USB hub/ethernet chip which provides the extra USB ports and the ethernet.

The USB part of the SoC has not changes between any Pi model - there are some well known issues with it, but in the vast majority of cases it works fine.
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Contrary to popular belief, humorous signatures are allowed. Here's an example...
"My grief counseller just died, luckily, he was so good, I didn't care."

user12345
Posts: 59
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Sun Mar 11, 2018 1:16 pm

It just suprise me that you ask for Software power values that are not accurate.
The LED never blinks, the error occurs too with a 3 A PSU. Its clear now that it is not a Power problem.
jamesh wrote:
Sat Mar 10, 2018 5:58 pm
Just to fill in some gaps.

There is ONE USB port of the SoC's used on Pi0,1,2,3. On the 1,2 and 3 this is plugged in to a USB hub/ethernet chip which provides the extra USB ports and the ethernet.

The USB part of the SoC has not changes between any Pi model - there are some well known issues with it, but in the vast majority of cases it works fine.
If there is no different HUB, and no revision changes in the Chips, and no interrupt from onboad wifi/BT, it means it is a Kernel Bug.
But what now, will it be fixed?
Would the PI2 run on a lower Kernel than PI3?

User avatar
FTrevorGowen
Forum Moderator
Forum Moderator
Posts: 4947
Joined: Mon Mar 04, 2013 6:12 pm
Location: Bristol, U.K.
Contact: Website

Re: RPI3 USB Hardware Bug?

Sun Mar 11, 2018 2:44 pm

Having re-read your first post and, in the light of subsequent discussion, I think that you are both over thinking and/or misinterpreting your issue. I do agree (now) that it's not necessarily a "power/current issue". I suspect your "Alpha Wlan Stick" is the problem (or rather it's drivers, which will have, almost certainly, been "provided" via Debian rather than the RPF). The output of lsusb which, I don't think you've posted (and that of usb-devices) will identify the chipset in use which may help others comment on any specific issues associated with such.
Trev.
Still running Raspbian Jessie on some older Pi's (an A, B1, B2, B+, P2B, 3xP0, P0W) but Stretch on my 2xP3A+, P3B+, P3B, B+, A+ and a B2. See: https://www.cpmspectrepi.uk/raspberry_pi/raspiidx.htm

user12345
Posts: 59
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Sat Mar 17, 2018 8:52 am

FTrevorGowen wrote:
Sun Mar 11, 2018 2:44 pm
I suspect your "Alpha Wlan Stick" is the problem (or rather it's drivers, which will have, almost certainly, been "provided" via Debian rather than the RPF). The output of lsusb which, I don't think you've posted (and that of usb-devices) will identify the chipset in use which may help others comment on any specific issues associated with such.
Trev.
Bus 001 Device 005: ID 148f:3070 Ralink Technology, Corp. RT2870/RT3070 Wireless Adapter
(rt2800usb, rt28070.bin Firmware 0.29)

T: Bus=01 Lev=02 Prnt=02 Port=03 Cnt=03 Dev#= 5 Spd=480 MxCh= 0
D: Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs= 1
P: Vendor=148f ProdID=3070 Rev=01.01
S: Manufacturer=Ralink
S: Product=802.11 n WLAN
S: SerialNumber=1.0
C: #Ifs= 1 Cfg#= 1 Atr=80 MxPwr=450mA
I: If#= 0 Alt= 0 #EPs= 7 Cls=ff(vend.) Sub=ff Prot=ff Driver=rt2800usb

But why the Alpha Stick, that works 24/7, years long, without any problem on RPI1 should be the problem?

user12345
Posts: 59
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Tue Mar 20, 2018 10:51 am

The Intresting things are that it happends on cold days much more frequency. That would speak for a Hardware Bug.
The other thing is "lsusb","usb-devices", "iwconfig" and "airmon-ng" give the same results if the device is dead(receive no Data) as it would be
still there.

User avatar
FTrevorGowen
Forum Moderator
Forum Moderator
Posts: 4947
Joined: Mon Mar 04, 2013 6:12 pm
Location: Bristol, U.K.
Contact: Website

Re: RPI3 USB Hardware Bug?

Tue Mar 20, 2018 3:37 pm

user12345 wrote:
Tue Mar 20, 2018 10:51 am
The Intresting things are that it happends on cold days much more frequency. That would speak for a Hardware Bug.
...
More likely a "hardware connector" issue.
Trev.
Still running Raspbian Jessie on some older Pi's (an A, B1, B2, B+, P2B, 3xP0, P0W) but Stretch on my 2xP3A+, P3B+, P3B, B+, A+ and a B2. See: https://www.cpmspectrepi.uk/raspberry_pi/raspiidx.htm

user12345
Posts: 59
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Thu Mar 22, 2018 3:02 pm

FTrevorGowen wrote:
Tue Mar 20, 2018 3:37 pm
More likely a "hardware connector" issue.
Trev.
Its not possible. Cause of 2 different USB Cables, cause of trying other USB Ports.
Bad soldering on more than 1 USB Port on one RPI3?!

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 22695
Joined: Sat Jul 30, 2011 7:41 pm

Re: RPI3 USB Hardware Bug?

Thu Mar 22, 2018 3:12 pm

user12345 wrote:
Thu Mar 22, 2018 3:02 pm
FTrevorGowen wrote:
Tue Mar 20, 2018 3:37 pm
More likely a "hardware connector" issue.
Trev.
Its not possible. Cause of 2 different USB Cables, cause of trying other USB Ports.
Bad soldering on more than 1 USB Port on one RPI3?!
I think that was the point being made, its could be a bad joint on the Pi PCB. In fact, it does sound like that is a good possibility, heating up makes the joint fail.
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Contrary to popular belief, humorous signatures are allowed. Here's an example...
"My grief counseller just died, luckily, he was so good, I didn't care."

user12345
Posts: 59
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Sat Mar 31, 2018 2:47 pm

jamesh wrote:
Thu Mar 22, 2018 3:12 pm
I think that was the point being made, its could be a bad joint on the Pi PCB. In fact, it does sound like that is a good possibility, heating up makes the joint fail.
Its impossible that USB-Device a) works 24/7 = Webcam and USB-Device B) fails = Wifi Stick.
The USB Ports are exchanged, the Wifi Stick+Cables+PSU exchanged.
It can be only a Hardware Design error(such one like exposure RPI3 to light and it crashes) or a Kernel Bug.

user12345
Posts: 59
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Tue Apr 03, 2018 9:35 pm

And the follow happend sometimes when the automatic Modprobe Restart fail. Its very annoying, modprobe is unable to flush the driver.
Then you try to manual reboot, but the PI hangs. You have the second manual invention plug off the RPI from Power and plug it in again.

Today we had alot Bluetooth traffic. Maybe the internal BT chip confuse the System? Even readed the PI3 can be hacked via onboard Wifi.


Apr 3 21:20:10 rpi3 kernel: [2941763.835071] usbcore: deregistering interface driver rt2800usb
Apr 3 21:20:11 rpi3 kernel: [2941764.125553] ieee80211 phy0: rt2x00queue_flush_queue: Warning - Queue 14 failed to flush
Apr 3 21:24:16 rpi3 kernel: [2942009.356974] INFO: task modprobe:21019 blocked for more than 120 seconds.
Apr 3 21:24:16 rpi3 kernel: [2942009.356986] Not tainted 4.9.52-v7+ #1038
Apr 3 21:24:16 rpi3 kernel: [2942009.356990] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Apr 3 21:24:16 rpi3 kernel: [2942009.356997] modprobe D 0 21019 21018 0x00000000
Apr 3 21:24:16 rpi3 kernel: [2942009.357045] [<80716460>] (__schedule) from [<80716984>] (schedule+0x50/0xa8)
Apr 3 21:24:16 rpi3 kernel: [2942009.357066] [<80716984>] (schedule) from [<8055c09c>] (usb_kill_urb+0xac/0xe0)
Apr 3 21:24:16 rpi3 kernel: [2942009.357105] [<8055c09c>] (usb_kill_urb) from [<7f280c48>] (rt2x00usb_uninitialize+0xac/0x134 [rt2x00usb])
Apr 3 21:24:16 rpi3 kernel: [2942009.357194] [<7f280c48>] (rt2x00usb_uninitialize [rt2x00usb]) from [<7f34eea4>] (rt2x00lib_uninitialize+0x58/0x74 [rt2x00lib])
Apr 3 21:24:16 rpi3 kernel: [2942009.357280] [<7f34eea4>] (rt2x00lib_uninitialize [rt2x00lib]) from [<7f34f1e4>] (rt2x00lib_remove_dev+0xa4/0x160 [rt2x00lib])
Apr 3 21:24:16 rpi3 kernel: [2942009.357339] [<7f34f1e4>] (rt2x00lib_remove_dev [rt2x00lib]) from [<7f2812a8>] (rt2x00usb_disconnect+0x28/0x50 [rt2x00usb])
Apr 3 21:24:16 rpi3 kernel: [2942009.357369] [<7f2812a8>] (rt2x00usb_disconnect [rt2x00usb]) from [<80560698>] (usb_unbind_interface+0x8c/0x2a0)
Apr 3 21:24:16 rpi3 kernel: [2942009.357397] [<80560698>] (usb_unbind_interface) from [<804fc838>] (__device_release_driver+0x94/0x118)
Apr 3 21:24:16 rpi3 kernel: [2942009.357417] [<804fc838>] (__device_release_driver) from [<804fc9f0>] (driver_detach+0xdc/0xe0)
Apr 3 21:24:16 rpi3 kernel: [2942009.357435] [<804fc9f0>] (driver_detach) from [<804fbb20>] (bus_remove_driver+0x5c/0xb0)
Apr 3 21:24:16 rpi3 kernel: [2942009.357455] [<804fbb20>] (bus_remove_driver) from [<804fce4c>] (driver_unregister+0x38/0x58)
Apr 3 21:24:16 rpi3 kernel: [2942009.357473] [<804fce4c>] (driver_unregister) from [<8055f9ec>] (usb_deregister+0x74/0xe8)
Apr 3 21:24:16 rpi3 kernel: [2942009.357500] [<8055f9ec>] (usb_deregister) from [<7f2d7410>] (rt2800usb_driver_exit+0x18/0x1c [rt2800usb])
Apr 3 21:24:16 rpi3 kernel: [2942009.357532] [<7f2d7410>] (rt2800usb_driver_exit [rt2800usb]) from [<801a1c70>] (SyS_delete_module+0x184/0x1d0)
Apr 3 21:24:16 rpi3 kernel: [2942009.357568] [<801a1c70>] (SyS_delete_module) from [<801080c0>] (ret_fast_syscall+0x0/0x1c)
Apr 3 21:26:19 rpi3 kernel: [2942132.237625] INFO: task modprobe:21019 blocked for more than 120 seconds.
Apr 3 21:26:19 rpi3 kernel: [2942132.237637] Not tainted 4.9.52-v7+ #1038
Apr 3 21:26:19 rpi3 kernel: [2942132.237642] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Apr 3 21:26:19 rpi3 kernel: [2942132.237648] modprobe D 0 21019 21018 0x00000000
Apr 3 21:26:19 rpi3 kernel: [2942132.237698] [<80716460>] (__schedule) from [<80716984>] (schedule+0x50/0xa8)
Apr 3 21:26:19 rpi3 kernel: [2942132.237718] [<80716984>] (schedule) from [<8055c09c>] (usb_kill_urb+0xac/0xe0)
Apr 3 21:26:19 rpi3 kernel: [2942132.237757] [<8055c09c>] (usb_kill_urb) from [<7f280c48>] (rt2x00usb_uninitialize+0xac/0x134 [rt2x00usb])
Apr 3 21:26:19 rpi3 kernel: [2942132.237829] [<7f280c48>] (rt2x00usb_uninitialize [rt2x00usb]) from [<7f34eea4>] (rt2x00lib_uninitialize+0x58/0x74 [rt2x00lib])
Apr 3 21:26:19 rpi3 kernel: [2942132.237912] [<7f34eea4>] (rt2x00lib_uninitialize [rt2x00lib]) from [<7f34f1e4>] (rt2x00lib_remove_dev+0xa4/0x160 [rt2x00lib])
Apr 3 21:26:19 rpi3 kernel: [2942132.237972] [<7f34f1e4>] (rt2x00lib_remove_dev [rt2x00lib]) from [<7f2812a8>] (rt2x00usb_disconnect+0x28/0x50 [rt2x00usb])
Apr 3 21:26:19 rpi3 kernel: [2942132.238002] [<7f2812a8>] (rt2x00usb_disconnect [rt2x00usb]) from [<80560698>] (usb_unbind_interface+0x8c/0x2a0)
Apr 3 21:26:19 rpi3 kernel: [2942132.238025] [<80560698>] (usb_unbind_interface) from [<804fc838>] (__device_release_driver+0x94/0x118)
Apr 3 21:26:19 rpi3 kernel: [2942132.238046] [<804fc838>] (__device_release_driver) from [<804fc9f0>] (driver_detach+0xdc/0xe0)
Apr 3 21:26:19 rpi3 kernel: [2942132.238063] [<804fc9f0>] (driver_detach) from [<804fbb20>] (bus_remove_driver+0x5c/0xb0)
Apr 3 21:26:19 rpi3 kernel: [2942132.238081] [<804fbb20>] (bus_remove_driver) from [<804fce4c>] (driver_unregister+0x38/0x58)
Apr 3 21:26:19 rpi3 kernel: [2942132.238099] [<804fce4c>] (driver_unregister) from [<8055f9ec>] (usb_deregister+0x74/0xe8)
Apr 3 21:26:19 rpi3 kernel: [2942132.238128] [<8055f9ec>] (usb_deregister) from [<7f2d7410>] (rt2800usb_driver_exit+0x18/0x1c [rt2800usb])
Apr 3 21:26:19 rpi3 kernel: [2942132.238156] [<7f2d7410>] (rt2800usb_driver_exit [rt2800usb]) from [<801a1c70>] (SyS_delete_module+0x184/0x1d0)
Apr 3 21:26:19 rpi3 kernel: [2942132.238179] [<801a1c70>] (SyS_delete_module) from [<801080c0>] (ret_fast_syscall+0x0/0x1c)
Apr 3 21:26:37 rpi3 shutdown[21266]: shutting down for system reboot
Apr 3 21:26:37 rpi3 init: Switching to runlevel: 6
Apr 3 21:26:38 rpi3 LCDd: sock_send: socket write error
Apr 3 21:26:38 rpi3 ifplugd(eth0)[1910]: Exiting.
Apr 3 21:26:40 rpi3 ntpd[2307]: ntpd exiting on signal 15

user12345
Posts: 59
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Tue Apr 17, 2018 12:19 pm

So, i tryed a 3rd USB Cable, this time a active one, same Problem.
I bought a RPI2, cause maybe the onboard wifi/bluetooth chip may interrupt something, same Problem.

The happend more often if it is cold was wrong, it simply increased over the time. We get more wifi traffic cause
german phone provider forced all people to use a Router->IP phone cause analog phone is canceled.

Cause RPI2 and RPI3 are a different Hardware Design the only thing thats stays is a long standing Kernel Bug.

user12345
Posts: 59
Joined: Mon Dec 22, 2014 7:06 pm

Re: RPI3 USB Hardware Bug?

Fri Apr 27, 2018 9:10 pm

4 USB Cables(even a 0.2 meter long dont help), 4 Powersupplys, 3 wifi sticks, 2 SD Cards, 2 RPI´s and no Bugfix.
I read the BananaPI support is bad but this outclass all.

Return to “Troubleshooting”