Raspberry Pi client crashes system with FTDI Device

Submitted by ove on #1

Dear Support,

I have a setup using:

- Server running on Synology DS212j DSM 6.0 registered with "3 device" license
- Attached FT232RL USB device (RFXtrx433)

- Client running on Raspberry Pi
- Client running in daemon mode - using the provided init script
- device connected via network and used as /dev/ttyUSB0

So far I was able to connect and use the /dev/ttyUSB0 device for a couple (5-10) seconds. Then the whole Raspberry Pi hangs - no network access / ui freezes. Reboot is needed on the client.

Log on the client:

Apr 16 10:38:19 home kernel: [ 570.539624] usb 2-1: new full-speed USB device number 3 using vhci_hcd
Apr 16 10:38:19 home kernel: [ 570.659540] usb 2-1: SetAddress Request (3) to port 0
Apr 16 10:38:20 home kernel: [ 570.765992] usb 2-1: New USB device found, idVendor=0403, idProduct=6001
Apr 16 10:38:20 home kernel: [ 570.766021] usb 2-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
Apr 16 10:38:20 home kernel: [ 570.766039] usb 2-1: Product: RFXtrx433
Apr 16 10:38:20 home kernel: [ 570.766056] usb 2-1: Manufacturer: RFXCOM
Apr 16 10:38:20 home kernel: [ 570.766072] usb 2-1: SerialNumber: A18RCK2
Apr 16 10:38:20 home kernel: [ 570.783844] ftdi_sio 2-1:1.0: FTDI USB Serial Device converter detected
Apr 16 10:38:20 home kernel: [ 570.784049] usb 2-1: Detected FT232RL
Apr 16 10:38:20 home kernel: [ 570.801373] usb 2-1: FTDI USB Serial Device converter now attached to ttyUSB0
Apr 16 10:40:55 home kernel: [ 725.871771] vhci_hcd: dequeue a urb b82f7f00
Apr 16 10:40:55 home kernel: [ 725.871808] vhci_hcd: device b9067990 seems to be still connected
Apr 16 10:40:55 home kernel: [ 725.872030] vhci_hcd: unlink->seqnum 54
Apr 16 10:40:55 home kernel: [ 725.872052] vhci_hcd: the urb (seqnum 54) was already given back
Apr 16 10:40:55 home kernel: [ 725.872294] vhci_hcd: dequeue a urb b82f7080
Apr 16 10:40:55 home kernel: [ 725.872315] vhci_hcd: device b9067990 seems to be still connected
Apr 16 10:40:55 home kernel: [ 725.872529] vhci_hcd: unlink->seqnum 55
Apr 16 10:40:55 home kernel: [ 725.872547] vhci_hcd: the urb (seqnum 55) was already given back
Apr 16 10:40:55 home kernel: [ 725.965873] vhci_hcd: dequeue a urb b82f7f00
Apr 16 10:40:55 home kernel: [ 725.966181] vhci_hcd: device b9067990 seems to be still connected
Apr 16 10:40:55 home kernel: [ 725.966361] vhci_hcd: unlink->seqnum 80
Apr 16 10:40:55 home kernel: [ 725.966379] vhci_hcd: the urb (seqnum 80) was already given back
Apr 16 10:40:55 home kernel: [ 725.966422] vhci_hcd: dequeue a urb b82f7080
Apr 16 10:40:55 home kernel: [ 725.966437] vhci_hcd: device b9067990 seems to be still connected
Apr 16 10:40:55 home kernel: [ 725.966612] vhci_hcd: unlink->seqnum 81
Apr 16 10:40:55 home kernel: [ 725.966629] vhci_hcd: the urb (seqnum 81) was already given back

Any ideas why the system might crash?

Best,
Ove

Submitted by Michael on Sun, 2016/04/17 - 11:08

#2

Make sure you are running the latest raspbian kernel like this

sudo apt-get update;sudo apt-get upgrade -y;sudo rpi-update;sync;sudo reboot

VirtualHere relies on the usbip client driver built into the raspbian kernel, it sounds like this probably has a bug. There was some fixes to the upstream kernel in 4.6 that went in to fix a possible lockup in the vhci_hcd driver inside the kernel. This is in mainline for 4.6 of the kernel, is not on the raspberry pi and 4.6rc3 has just been released. So you might have to wait until 4.6 is in the raspbian kernel. I noticed a few days ago they updated the raspbian kernel to 4.4, anyway the patch is here and you might like to recompile raspbian with the patch

https://github.com/torvalds/linux/commit/21619792d1eca7e772ca190ba68588…