hplip setup problems

[Isaac Mushinsky]

On Wed, Mar 26, 2008 at 11:02 AM, Isaac Mushinsky <itz at mushinsky.net> wrote:

> I've tried freebsd-questions, but to no avail so far, so please excuse my
> posting here in the hope to find people more familiar with this...
>
> I am having trouble setting up HP Photosmart C4280 all-in-one device for
> printing. hpssd, cups are started in the correct order. Device node
> permissions seem to be correct too.
>
> 1. This is an all-in-one device. (I did not try the scanner setup yet)
> 2. ulpt, umass, uscanner modules are not in kernel and not kld-loaded.
> 3. Machine is FreeBSD 7-stable/amd64.
> 4. hplip is 2.8.2
>
> $ usbdevs
> addr 1: UHCI root hub, Intel
> addr 1: UHCI root hub, Intel
> addr 1: UHCI root hub, Intel
> addr 1: EHCI root hub, Intel
>  addr 2: Photosmart C4200 series, HP
> addr 1: UHCI root hub, Intel
> addr 1: UHCI root hub, Intel
> addr 1: UHCI root hub, Intel
> addr 1: EHCI root hub, Intel
> There is a ppd file for these series with the hplip distribution.
> hpssd, cups started in the order needed. hp-setup detects the printer, and
> then says 'Unable to create queue'. CUPS web interface actually adds the
> printer, but then cannot print test page to it. cups user runs hpssd (I
> tried root too).
> /var/log/messages has:
> Mar 25 23:57:56 omsk python: io/hpmud/musb.c 135: unable
> get_string_descriptor -5: Input/output error
> Mar 25 23:57:56 omsk python: io/hpmud/musb.c 1951: invalid serial id
> string ret=-5
>
>
> With USB_DEBUG=4 (for libusb), the hp-setup  fails like this:
>
> usb_os_find_devices: Found /dev/ugen0 on /dev/usb3
> usb_control_msg: 128 6 512 0 0x7fffffff9ed0 8 1000
> usb_control_msg: 128 6 512 0 0x80131d080 124 1000
> usb_control_msg: 128 6 770 1033 0x7fffffff9e40 255 5000
> usb_control_msg: 128 6 771 1033 0x7fffffff9e40 255 5000
> USB error: error sending control message: Input/output error
> ...
> device `hpaio:/usb/Photosmart_C4200_series?serial=0' is a Hewlett-Packard
> Photosmart_C4200_series all-in-one
>
> This 'serial=0' seems to be the result of failed usb_control_msg call.
>
>  I exhibited here all relevant output (/var/log/messages, usbdevs, ls -l
> /dev etc.):
> http://omsk.mushinsky.net/hplip-trouble
>
> Any help is appreciated.
> -isaac-
>
>

What is this serial id string? is this supposed to be a unique string
assigned to each produced device by the vendor? could it be that it is just
HP's fault, or the serial id is on a wrong node of a multi-device somehow? I
thought the only use for it is to uniquely identify a device, since I only
have one, can I just patch libusb to fake some string?