[Unresolved] Xsane scanner tool

mark2 · Posted: Thu Oct 08, 2015 2:50 am Post subject:

Thanks, russK. That produced quite an output, around 23000 lines, but here it is: https://bpaste.net/show/ca1a354cfc75

But the first part gave:

russK · l33t Joined: 27 Jun 2006 Posts: 665

A few things I noticed that would be worth looking into, unfortunately some could be red herrings:

line 12980 and several other places:

mark2 · Posted: Thu Oct 08, 2015 5:46 am Post subject:

Ok, I don't have the hp backend in the make.conf, but if it won't hurt anything, I'll try your suggestion with that, though I suspect that won't really make any difference.

I will check into the udev rules for /dev/scanner.

The change on the device seems to have taken place because I did disconnect the usb cable from the back of the scanner itself, but I put it right back to test a suggestion I had read. So it was not like I moved it to a different usb port on the computer - THAT did not change. I wondered about the 008 / 003 myself as it didn't make sense. It was definitely timing out, though.

I got sane-find-scanner to show this:

mark2 · Posted: Thu Oct 08, 2015 7:13 am Post subject:

I re-configured the kernel to make sure that OHCI and EHCI were compiled as modules. I could not find CONFIG_CHR_DEV_SG anywhere in the kernel configuration. My kernel version is 4.0.5, so maybe it's just not in there.

I did add the hp backend in make.conf.

I wrote a rule in /etc/udev/rules.d/91-local.rules that I found in searching and it looks like this:

mark2 · Posted: Thu Oct 08, 2015 7:37 am Post subject:

Update: I did a reboot, ran xsane and it reported the device 008 / 002 this time, but it still cannot open avision:008:003, Operation not supported.

I left it running emerge --sync since it said that it was last run 31 days ago.
_________________
Thanks,
Mark
__________________________________________________________________
"Flying is learning how to throw yourself at the ground and miss." - Douglas Adams

russK · l33t Joined: 27 Jun 2006 Posts: 665

bbgermany · Posted: Thu Oct 08, 2015 1:11 pm Post subject:

mark2 · Posted: Fri Oct 09, 2015 1:14 am Post subject:

@russK:

I went back through the kernel and did find Device Drivers -> SCSI device support -> SCSI generic support. It was set as a module and was like that from the get-go. So, just to test, I switched to the kernel, and recompiled. Now it's running emerge -auvDN @world and it is doing about 46 packages, most of them updates.

I feel like there is something I'm supposed to do after it finishes, just not sure what that might be. Depclean? I don't know....

@bb:

I may made a typo in reporting 008 /003. For now Xsane, scanimage -L, and sane-find-scanner all report the device 008:002. Sorry, it's late and I'm just not positive on that note.
_________________
Thanks,
Mark
__________________________________________________________________
"Flying is learning how to throw yourself at the ground and miss." - Douglas Adams

russK · l33t Joined: 27 Jun 2006 Posts: 665

russK · l33t Joined: 27 Jun 2006 Posts: 665

Oh sorry about the hplip suggestion, I missed the fact that hplip doesn't support scanners that don't print :oops:

mark2 · Posted: Fri Oct 09, 2015 7:35 am Post subject:

I did try the hplip with no luck, of course. This one has a button that can send to a printer, but interestingly none of the buttons on the scanner will work or initiate even a manual scan. There may very well be something dead in it. But anyway, last ditch effort:

dmesg after reconnecting the scanner can be seen here: https://bpaste.net/show/a1186da99a46

I ran xsane after that and this time it referred to 008 / 003, saying:

bbgermany · Posted: Fri Oct 09, 2015 9:17 am Post subject:

I have an old Canoscan Lide 20 laying in the closet. Ill try to figure out how to get this running, maybe we can adopt it to your avision config then

greets, bb

EDIT: I looked a bit around the manpages for sane-backends and I have a few questions:
1st: Do you use libusb? check if /proc/bus/usb is mounted
2nd: Could you replace the line "usb libusb:008:003" with "usb 0x03f0 0x0b01" instead and try again?
3rd: Do you have sane-backends compiled with debug? If yes, you could use "export SANE_DEBUG_AVISION=7" for more information about the issue.
4th: A studid question, but is saned running?
_________________
Desktop: Ryzen 5 5600G, 32GB, 2TB, RX7600
Notebook: Dell XPS 13 9370, 16GB, 1TB
Server #1: Ryzen 5 Pro 4650G, 64GB, 16.5TB
Server #2: Ryzen 4800H, 32GB, 22TB

mark2 · Posted: Sat Oct 10, 2015 1:14 am Post subject:

mark2 · Posted: Sat Oct 10, 2015 1:20 am Post subject:

Update: emerge sane-backends --debug failed to emerge media-gfx/sane-backends-1.0.24-r5.

The debug info is here: https://bpaste.net/show/0db5e7bb079e

That's over 16,000 lines and I don't have the experience to make sense out of that.
_________________
Thanks,
Mark
__________________________________________________________________
"Flying is learning how to throw yourself at the ground and miss." - Douglas Adams

russK · l33t Joined: 27 Jun 2006 Posts: 665

mark2 · Posted: Sat Oct 10, 2015 1:57 pm Post subject:

First, the output of your suggested scanimage produced this: https://bpaste.net/show/1ce0c459c485

Interesting. So it wouldn't let me emerge sane-backends-1.0.24-r5 when I tried. I dug into /usr/portage/elog by using a terminal on elog, which btw had a big red X over it and kept saying permissions denied. But in the terminal I could navigate into media-gfx/sane-backends to see what was actually in there. And I found some files that may be causing the problem here at: https://bpaste.net/show/cda6027a413f

And another folder called 'files'. The contents of it are here: https://bpaste.net/show/5ba516f50802

So I am wondering if there might be something in editing the saned.confd. It is showing this:

mark2 · Posted: Sat Oct 10, 2015 6:20 pm Post subject:

So I thought that if the C compiler could not run executables, then I should switch to root and emerge gcc. It tried to emerge gcc-4.9.3, but that also failed with the same error message about executables.

Looks like a stalemate. I did find:

russK · l33t Joined: 27 Jun 2006 Posts: 665

russK · l33t Joined: 27 Jun 2006 Posts: 665

Another possibility, I noticed here on the scanimage manpage, 'scanimage -A' produces some help output with a lot of options, and for mine I saw this:

mark2 · Posted: Sat Oct 10, 2015 8:46 pm Post subject:

Well, I had already managed to upgrade to gcc-4.9.3, but trying scanimage -A and scanimage --source Flatbed produced the same error as before. open of device avision:libusb:001:019 failed: Operation not supported.

It is indicative of something that when I launch xsane as root, the scanner buttons light up, but nothing else happens. Also true of the other scanner commands. So something triggers activity, but it does not go anywhere.

The upgrade did allow me to emerge sane-backends, though.
_________________
Thanks,
Mark
__________________________________________________________________
"Flying is learning how to throw yourself at the ground and miss." - Douglas Adams

sebB · l33t Joined: 02 Mar 2011 Posts: 806 Location: S.O. France

After adding SANE_BACKENDS="avision" to your make.conf, have you recompiled sane-backends?

russK · l33t Joined: 27 Jun 2006 Posts: 665

Another shot in the dark might be to try the sane-backends 1.0.25_pre20150628 ebuild or the live 9999 ebuild.

mark2 · Posted: Sun Oct 11, 2015 12:09 am Post subject:

@ sebB:

Yes.

@: russK:

I tried emerge sane-backends 1.0.25_pre20150628, but it says no ebuilds to satisfy. How do I get it to do that? Right now it has done 1.0.24-r4, but when I run the gui it shows 0.999 in the title bar.
_________________
Thanks,
Mark
__________________________________________________________________
"Flying is learning how to throw yourself at the ground and miss." - Douglas Adams

russK · l33t Joined: 27 Jun 2006 Posts: 665

mark,

maybe I typed it wrong, you can see all of the ebuilds in portage like this with 'equery', which is in the gentoolkit package: