[sane-devel] scaner on HP LaserJet M1005 MFP

[m. allan noah]

On Wed, Apr 7, 2010 at 1:14 PM, viktor vraniak <tyctor at post.cz> wrote:
>> > so i think, opened device is renewed and scanimage lost/not get descriptor for
>> new device?
>>
>> Yes- It appears that the device is reset (either by the kernel or by
>> itself), after this attempted write. The reset succeeds, and so the
>> device is found by the second call to scanimage.
>>
>> does a long-running frontend like xsane exhibit this behaviour? What
>> about the -b argument to scanimage (should make multiple scans)
>>
>> allan
>> --
>> "The truth is an offense, but not a sin"
>>
> sorry again bad address... :o(
>
> i have tried
> scanimage -b - it scans only first page, and hangs on second (only kill proces helps)
> xscanimage - if i acquire preview, i cannot scan, programs hangs (only kill helps)
>                     - if i directly scan, i can scan only once, second scan hangs program (only kill helps)
> same as xscanimage is xsane
>
> so i have to find what causes this device USB reset.
> how to do this?
> can you help me with this?
>

This would imply that something the backend is doing at the end of a
scan is confusing the scanner or kernel. Can you try this with a
different computer, or older kernel version, or other USB cable or
something? If none of those things help, then you would have to try
and compare the packets sent by the backend to those sent by the
windows driver, using a tool like this:

http://www.pcausa.com/Utilities/UsbSnoop/

You might also try to contact the backend author directly.

allan
-- 
"The truth is an offense, but not a sin"