Hi Nicolas, <br><br>Yes we ran through some debug and I think reached the point where we were both out of ideas :)<br><br>I was sort-of hoping that this latest pixma issue could be related, but I think probably not....<br>
<br>I think I have to do the USB-protocol level analysis, as I don't think this is an install issue, but I've not had much free time recently.<br><br>Also the windows USB-snoop applications don't dump in a format that is suitable for diffing with the Linux usbmon output, so I've been trying to understand how other developers perform this analysis. It seems to me that manual analysis is time-consuming and error-prone.<br>
<br>A script in the experimental tree was mentioned to clean up the usbsniff output, so I will give that a try next.<br><br>Thanks,<br><br>Steve<br><br><br><br><div class="gmail_quote">2009/2/28 Nicolas Martin <span dir="ltr"><<a href="mailto:nicolas.martin@freesurf.fr">nicolas.martin@freesurf.fr</a>></span><br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">This lead to some direct exchanges between us a while back, but it looks<br>
to me more like some low level usb issue between the MF4660 scanner and<br>
libusb ?<br>
The scanner stops responding at some point, but all the pixma backend<br>
messages exchanged look identical to those in the Win Snoop.<br>
So don't understand yet what can be the fail reason.<br>
If you can have a look at a usb port monitoring, that may probably help.<br>
<br>
Nicolas<br>
<br>
Le jeudi 26 février 2009 à 18:13 +0000, Steve Hardy a écrit :<br>
<div><div></div><div class="Wj3C7c">> Ok, thanks for the clarification, I discovered that there was a<br>
> duplicate /etc/sane.d directory in /usr/etc. I ran the sane "make<br>
> install" to overwrite my installed sane library, which probably<br>
> explains it - I have now removed the original sane package completely,<br>
> then reinstalled from CVS and my scanner still hangs up.<br>
><br>
> The steps I see are :<br>
><br>
> 1 - Start scanimage -T<br>
> 2 - Scan head moves to the "start scan" position<br>
> 3 - Scan head cycles through colours (light on/off)<br>
> 4 - Scan head starts scan pass, but then stops about 1/3 of the way<br>
> accoss the page.<br>
> 5 - Scan head returns to "start scan" position<br>
> 6 - Scanner is unresponsive (returns -ETIMEDOUT to everything) until<br>
> USB cable is removed and replaced or scanner is power cycled.<br>
><br>
> I have taken usb sniffs from windows and I can't spot any significant<br>
> differences, but comparing the usbsniff format with the usbmon capture<br>
> is pretty difficult.<br>
><br>
> When I have more time I plan to hack the usbsniff code so it will<br>
> output in a format that's diffable with usbmon, and put some more<br>
> debug into scanimage to try and spot where the problem occurs.<br>
><br>
><br>
><br>
><br>
> 2009/2/25 Gleb Baryshev <<a href="mailto:gleb.baryshev@gmail.com">gleb.baryshev@gmail.com</a>><br>
> Steve Hardy wrote:<br>
> > I have similar problems to this with the pixma backend and a<br>
> Canon<br>
> > MF4660 - only my MFP *always* returns the -ETIMEOUT response<br>
> after<br>
> > moving the scan head to the end of the panel in preparation<br>
> for the scan.<br>
><br>
><br>
> I can't yet define exactly what actions hang up my scanner. At<br>
> least,<br>
> when I scan continuously, using the same program, all work<br>
> well.<br>
> By the way, successful scanning adds three lines like this to<br>
> the kernel<br>
> ring buffer:<br>
> usb 1-1: usbfs: interface 1 claimed by usblp while 'xsane'<br>
> sets config #1<br>
> (or 'scanimage' instead of 'xsane')<br>
><br>
> Dennis Lou wrote:<br>
> > If you keep reading, Gleb solved his problem by manually<br>
> removing his old version of sane prior to re-installing a<br>
> fresh CVS version (apparently automated package removal wasn't<br>
> doing a complete job). Perhaps that will also solve your<br>
> problem being that you two are experiencing similar things?<br>
><br>
><br>
> Actually, removing of old files was aimed to solve a problem<br>
> when<br>
> scanimage and xsane could only be run<br>
> from /usr/local/etc/sane.d (i.e.<br>
> scanimage couldn't find the scanner, but after 'cd<br>
> /usr/local/etc/sane.d' it could). At first it helped, but<br>
> after reboot<br>
> only xsane could run from anywhere. This isn't a big problem,<br>
> but the<br>
> fact remains.<br>
><br>
><br>
> Gleb<br>
><br>
> --<br>
> sane-devel mailing list: <a href="mailto:sane-devel@lists.alioth.debian.org">sane-devel@lists.alioth.debian.org</a><br>
> <a href="http://lists.alioth.debian.org/mailman/listinfo/sane-devel" target="_blank">http://lists.alioth.debian.org/mailman/listinfo/sane-devel</a><br>
> Unsubscribe: Send mail with subject "unsubscribe<br>
> your_password"<br>
> to <a href="mailto:sane-devel-request@lists.alioth.debian.org">sane-devel-request@lists.alioth.debian.org</a><br>
><br>
><br>
><br>
><br>
> --<br>
> Steve Hardy<br>
><br>
> Applied Realtime Systems Limited<br>
> <a href="http://www.appliedrealtime.co.uk" target="_blank">www.appliedrealtime.co.uk</a><br>
> Tel : 0845 8678251<br>
> Fax : 0845 8678241<br>
</div></div>> --<br>
<div><div></div><div class="Wj3C7c">> sane-devel mailing list: <a href="mailto:sane-devel@lists.alioth.debian.org">sane-devel@lists.alioth.debian.org</a><br>
> <a href="http://lists.alioth.debian.org/mailman/listinfo/sane-devel" target="_blank">http://lists.alioth.debian.org/mailman/listinfo/sane-devel</a><br>
> Unsubscribe: Send mail with subject "unsubscribe your_password"<br>
> to <a href="mailto:sane-devel-request@lists.alioth.debian.org">sane-devel-request@lists.alioth.debian.org</a><br>
<br>
</div></div></blockquote></div><br>