Disapearing pl2303 usb serial adapter on rpi2

[bob prohaska]

> 
> On Thu, Dec 21, 2017 at 05:29:05PM +0100, Hans Petter Selasky wrote:
> > Further, you might want to only enable USB HUB debug messages to figure 
> > out why the USB stack thinks the device is going away:
> > 
> > sysctl hw.usb.uhub.debug=17
> > 
> 

Tried it here, using usbconfig to power cycle the adapter. Console output
follows. The pl2303 is at unit 0 address 5. Console output follows:
[initial state is pl2303 stuck, no /dev/cuaU0 present, command is
usbconfig -u 0 -a 5 power_off]
 
uhub_read_port_status: port 3, wPortStatus=0x0100, wPortChange=0x0000, err=USB_ERR_NORMAL_COMPLETION
uhub_read_port_status: port 5, wPortStatus=0x0101, wPortChange=0x0000, err=USB_ERR_NORMAL_COMPLETION
usb_needs_explore: 
usb_bus_powerd: bus=0xd7e65000
uhub_explore: udev=0xd7daf000 addr=1
uhub_read_port_status: port 1, wPortStatus=0x0503, wPortChange=0x0000, err=USB_ERR_NORMAL_COMPLETION
uhub_explore: udev=0xd7db0000 addr=2
uhub_read_port_status: port 1, wPortStatus=0x0503, wPortChange=0x0000, err=USB_ERR_NORMAL_COMPLETION
uhub_read_port_status: port 2, wPortStatus=0x0503, wPortChange=0x0000, err=USB_ERR_NORMAL_COMPLETION
uhub_read_port_status: port 3, wPortStatus=0x0100, wPortChange=0x0000, err=USB_ERR_NORMAL_COMPLETION
uhub_read_port_status: port 4, wPortStatus=0x0103, wPortChange=0x0000, err=USB_ERR_NORMAL_COMPLETION
uhub_explore: udev=0xd8cdb000 addr=6
uhub_read_port_status: port 1, wPortStatus=0x0103, wPortChange=0x0000, err=USB_ERR_NORMAL_COMPLETION
uhub_read_port_status: port 2, wPortStatus=0x0303, wPortChange=0x0000, err=USB_ERR_NORMAL_COMPLETION
uhub_read_port_status: port 3, wPortStatus=0x0100, wPortChange=0x0000, err=USB_ERR_NORMAL_COMPLETION
uhub_read_port_status: port 5, wPortStatus=0x0101, wPortChange=0x0000, err=USB_ERR_NORMAL_COMPLETION
uhub_explore_handle_re_enumerate: Unconfigure failed: USB_ERR_STALLED: Ignored.
uhub_reset_tt_callback: TT buffer reset
uhub_reset_tt_callback: TT buffer reset
usbd_req_re_enumerate: addr=5, set address failed! (USB_ERR_IOERROR, ignored)
uhub_reset_tt_callback: TT buffer reset
uhub_reset_tt_callback: TT buffer reset
usbd_setup_device_desc: getting device descriptor at addr 5 failed, USB_ERR_IOERROR
uhub_reset_tt_callback: TT buffer reset
uhub_reset_tt_callback: TT buffer reset
usbd_req_re_enumerate: addr=5, set address failed! (USB_ERR_IOERROR, ignored)
uhub_reset_tt_callback: TT buffer reset
uhub_reset_tt_callback: TT buffer reset
usbd_setup_device_desc: getting device descriptor at addr 5 failed, USB_ERR_IOERROR
usb_needs_explore: 
usb_bus_powerd: bus=0xd7e65000
uhub_explore: udev=0xd7daf000 addr=1
uhub_read_port_status: port 1, wPortStatus=0x0503, wPortChange=0x0000, err=USB_ERR_NORMAL_COMPLETION
uhub_explore: udev=0xd7db0000 addr=2
uhub_read_port_status: port 1, wPortStatus=0x0503, wPortChange=0x0000, err=USB_ERR_NORMAL_COMPLETION
uhub_read_port_status: port 2, wPortStatus=0x0503, wPortChange=0x0000, err=USB_ERR_NORMAL_COMPLETION
uhub_read_port_status: port 3, wPortStatus=0x0100, wPortChange=0x0000, err=USB_ERR_NORMAL_COMPLETION
uhub_read_port_status: port 4, wPortStatus=0x0103, wPortChange=0x0000, err=USB_ERR_NORMAL_COMPLETION
uhub_explore: udev=0xd8cdb000 addr=6
uhub_read_port_status: port 1, wPortStatus=0x0103, wPortChange=0x0000, err=USB_ERR_NORMAL_COMPLETION
uhub_read_port_status: port 2, wPortStatus=0x0303, wPortChange=0x0000, err=USB_ERR_NORMAL_COMPLETION
uhub_read_port_status: port 3, wPortStatu

Couldn't establish the timing exactly, but the resets correspond to about the
time of the power cycling. After power cycling there was no /dev/cuaU0 created.
Curiously, the power cycling _didn't_ trigger the error messages I saw earlier
on the controlling terminal; it appeared that the power cycling was successful,
even though /dev/cuaU0 wasn't created.

Again, unplugging and replugging the USB connector had no effect. Disconnecting
the TX, RX and ground and _then_ unplugging/replugging the USB end produced a
successful resest of the adapter. Singly disconnecting TX, RX _or_ ground did
nothing. Disconnecting both ends completely caused USB recognition and restored
normal operation.


Thanks for reading, and any ideas!

bob prohaska