[solved] (x)sane can't find any devices since 2.6.38.4 install [solved]

[Almost-retired]

I just discovered that the /etc/sane.d/dll.conf was only 8 bytes long AFTER adding the Epson 1250-u, but that the rpmsave versions were a bit over a kilobyte.

So I mv'd the longest one back to dll.conf, and everything is back among the living here.  In retrospect, Duh!!!

However [root@coyote xinetd.d]# /usr/bin/sane-find-scanner

  # sane-find-scanner will now attempt to detect your scanner. If the
  # result is different from what you expected, first make sure your
  # scanner is powered up and properly connected to your computer.

  # No SCSI scanners found. If you expected something different, make sure that
  # you have loaded a kernel SCSI driver for your SCSI adapter.

found USB scanner (vendor=0x04b8, product=0x0856) at libusb:001:004
found USB scanner (vendor=0x0403 [FTDI], product=0x6001 [USB FAST SERIAL ADAPTER]) at libusb:001:011
found USB scanner (vendor=0x0403 [FTDI], product=0x6001 [USB HS SERIAL CONVERTER], chip=GT-6816?) at libusb:001:019
found USB scanner (vendor=0x046d, product=0x08a2) at libusb:002:002
found USB scanner (vendor=0x04b8 [EPSON], product=0x010f [EPSON Scanner 010F], chip=LM9832/3) at libusb:002:004

There are 2 scanners present on this system.  Back at about 2.6.38 final, it was even showing me my pcHDTV-3000 card as a potential src.  Never tried it though.

I have been through some of the google-able checklists without any luck.

Any clues on how to re-enable my use of them?

Or is this the same fix I had to use to bring my pcHDTV-3000 back to life, change, in every src file of dvb-atsc-tools, the line #include linux/videodev.h to be videodev2.h, at which point it built, installed, worked, and so does kaffiene now.

Thanks & Cheers, Gene

[Almost-retired]

I have an email saying there is a reply here, but it is not being shown when I click on the link, or refresh.

[Ertain]

Hold on.  First of all, what are the scanners that aren't being detected?  I might be able to give some insight but I cannot guarantee anything.

[Almost-retired]

See the list, clips from /usr/bin/sane-find-scanner:

The first one is the scanner that is built into an Epson NX515.  The printer FWIW works normally.
That is this one in the find-scanner output
found USB scanner (vendor=0x04b8, product=0x0856) at libusb:001:004

Then I just noticed this 2nd one is different vendor/product:
found USB scanner (vendor=0x046d, product=0x08a2) at libusb:002:002
And its possible that may be a webcam, and old logitek...  Yes it is according to 'cheese'.

The the 3rd one is an older Epson 1250-u, this one:
found USB scanner (vendor=0x04b8 [EPSON], product=0x010f [EPSON Scanner 010F], chip=LM9832/3) at libusb:002:004

A 4th choice that also did show up in the xsane 'pick device' menu is:
01:08.0 Multimedia video controller: Conexant Systems, Inc. CX23880/1/2/3 PCI Video and Audio Decoder (rev 05)
01:08.2 Multimedia controller: Conexant Systems, Inc. CX23880/1/2/3 PCI Video and Audio Decoder [MPEG Port] (rev 05)
Which is a pcHDTV-3000 card in a pci slot.
But xsane can't find any of them.

Hints gleefully checked out.  Bearing in mind that I had to change all references to #include linux/videodev.h, in all the .c files of the tarball dvb-atsc-tools-1.0.7 from the pchdtv. site to videodev2.h, remake and reinstall before that tv card above would work again.  I've already emailed Jack the updated 1.0.8 tarball, so something should show up on his site in due time.

So I am suspecting it is something to do with the transition from V4L to V4L2 that is taking place in the 2.6.38+ time frame.

If you can help, be my guest, I know just enough C to be dangerous & have forgotten much of that over the last 20-25 years.

Thanks & Cheers, Gene.

[gseaman]

I just got updates from the repo, http://ftp.jaist.ac.jp/pub/Linux/PCLinuxOS/apt/, that included some change in packaging which now includes a new package, task-scanning. It did not get automatically installed, even though I have an all in one scanner. This may have something you need.

Galen

[Almost-retired]

task-scanning was a meta-pkg, and did not pull in anything I didn't already have.

Sorry it didn't work.

Thanks Galen, & Cheers, Gene

[gseaman]

Have you tried to reboot into one of the earlier kernels to see if it works there?

Galen

[Almost-retired]

Humm, it did work, about 3 weeks back, before I had a boot drive trashed by the seagate firmware update utility, and had to re-install.  I just went backwards and tried 2.6.38.2 and 2.6.33.7, the only other choices I have ATM, both _did_ work, now they don't.

So the puzzle gets a bit more difficult.  Any idea where I should start now that we know its a general failure?

Thanks Galen, & Cheers, Gene

[gseaman]

So, did it work after the firmware update on the new install, before the kernel update?

Galen

[Almost-retired]

I didn't have the need to check it then as I was concentrating more on bringing some of my own scripts back to life after what was a bare metal install, so I can't testify about that.  I do know it all worked quite well before the seagate firmware updater decided to not only scramble the BLKID's of my partitions, it also, without actually destroying the data, interchanged the labels I had also installed.  That IMNSHO, puts the big lie to the dependability of using the BLKID or a label to mount a partition I guess.  Not any better than using /dev/sdd3 to define /usr.

All the sane related stuff has been re-installed in the last 18 hours too.

Is there someplace I can set a higher debug level so I can see what its doing when I run it from a konsole?  Or is there a way to make strace trace the sub-processes as they are called?

Thanks & Cheers Galen, but its getting sleepy out as I'm just west of the little pond here, so I'll be back in the morning.

[gseaman]

I was hoping to identify what was the last thing you did before the xsane went awol. It may not have been the kernel.

I'm over on the west coast, so I have a couple of extra hours.

Galen

[Almost-retired]

I may be blaming the wrong thing.  And I think the logging sucks dead toads thru soda straws.  In /var/log/rpmpkgs and rpmpkgs.1 there is a list of pkgs, but it is both alphabetically sorted, and undated.

I always set my package manager to save the upgrades it downloads, and I find that I have at least 5 sane-related pkgs in /var/cache/apt/archives, all dated March 4th.  It is entirely possible that I hadn't needed to scan anything since they were installed.  So, lets use that as the next attack vector, by asking how I can make synaptic show, and re-install the old versions of
-rw-r--r-- 1 root root    72140 Apr 21 15:11 libsane-hpaio-3.11.3a-1pclos2011.i586.rpm
-rw-rw-r-- 1 root root   396504 Mar  4 15:54 sane-backends-1.0.21-4pclos2011.i586.rpm
-rw-rw-r-- 1 root root   395551 Mar  4 15:54 sane-backends-doc-1.0.21-4pclos2011.i586.rpm
-rw-rw-r-- 1 root root   123348 Mar  4 15:54 sane-backends-iscan-1.0.21-4pclos2011.i586.rpm
-rw-rw-r-- 1 root root    38316 Mar  4 15:54 saned-1.0.21-4pclos2011.i586.rpm

I think the first one likely has nothing to do with this, but just in case.
I found just now that PCC could add the old Perfection 1250, and that xsane now seems to see it.  Asking for a preview scan also seemed to work, after the prolonged lamp warmup.
However, PCC's scanner database contains none of the Epson MFP devices, and the scanner in the NX-515 has a far flatter illumination field than the much older Epson 1250-u.

To refresh, sane-find-scanner reports it just fine:

found USB scanner (vendor=0x04b8, product=0x0856) at libusb:001:004  <-the NX-515 BUT ONLY this disappears when I turn the printers power off

found USB scanner (vendor=0x0403 [FTDI], product=0x6001 [USB FAST SERIAL ADAPTER]) at libusb:001:011
found USB scanner (vendor=0x0403 [FTDI], product=0x6001 [USB HS SERIAL CONVERTER], chip=GT-6816?) at libusb:001:019
NDI what these two are.

found USB scanner (vendor=0x046d, product=0x08a2) at libusb:002:002 <-Logitech, Inc. Labtec Webcam Pro, a POS if there ever was one.

found USB scanner (vendor=0x04b8 [EPSON], product=0x010f [EPSON Scanner 010F], chip=LM9832/3) at libusb:002:004 <-the Perfection 1250-u

My playing around with the NX-515's power caused this in the dmesg log

usb 1-4: usbfs: interface 2 claimed by usb-storage while 'usb' sets config #1
usb 1-1.1.4.4.3: usbfs: interface 0 claimed by ftdi_sio while 'sane-find-scann' sets config #1
usb 1-1.1.4.4.3: usbfs: interface 0 claimed by ftdi_sio while 'sane-find-scann' sets config #1
usb 1-1.1.4.4.3: usbfs: interface 0 claimed by ftdi_sio while 'sane-find-scann' sets config #1
usb 1-1.1.4.4.3: usbfs: interface 0 claimed by ftdi_sio while 'sane-find-scann' sets config #1
usb 1-1.1.4.4.3: usbfs: interface 0 claimed by ftdi_sio while 'sane-find-scann' sets config #1
usb 1-1.1.4.4.3: usbfs: interface 0 claimed by ftdi_sio while 'sane-find-scann' sets config #1
usb 1-4: USB disconnect, address 4
usb 1-1.1.4.4.3: usbfs: interface 0 claimed by ftdi_sio while 'sane-find-scann' sets config #1
usb 1-4: new high speed USB device using ehci_hcd and address 20
usb 1-4: New USB device found, idVendor=04b8, idProduct=0856
usb 1-4: New USB device strings: Mfr=1, Product=2, SerialNumber=3
usb 1-4: Product: USB2.0 MFP(Hi-Speed)
usb 1-4: Manufacturer: EPSON
usb 1-4: SerialNumber: 4C454C593037313524
scsi9 : usb-storage 1-4:1.2
scsi 9:0:0:0: Direct-Access     EPSON    Storage          1.00 PQ: 0 ANSI: 2
sd 9:0:0:0: Attached scsi generic sg5 type 0
sd 9:0:0:0: [sde] Attached SCSI removable disk
usb 1-4: usbfs: interface 2 claimed by usb-storage while 'usb' sets config #1
usb 1-1.1.4.4.3: usbfs: interface 0 claimed by ftdi_sio while 'sane-find-scann' sets config #1

This thing also contains a card reader, called sg5 above, but its has never had a card plugged in as I always use a cable to the camera so static doesn't kill the card when I move it.
The point is that the scanner, IIRC used to be separately discovered.

Now I was going to look at the discovery in the numerous test reboots I did yesterday, and had the thought that I should grep the messages log for sane & got this:
Apr 18 04:02:42 localhost msec: tcp        0      0 *:sane-port                 *:*                         LISTEN      1987/xinetd
Apr 18 04:18:03 localhost synaptic: [RPM] libsane1-1.0.21-1pclos2010 removed
Apr 18 04:18:05 localhost synaptic: [RPM] sane-backends-1.0.21-1pclos2010 removed
Apr 18 04:19:27 localhost synaptic: [RPM] sane-backends-iscan-1.0.21-4pclos2011 installed
Apr 18 04:19:37 localhost xinetd[1987]: Server /usr/sbin/saned is not executable [file=/etc/xinetd.d/saned] [line=14]
Apr 18 04:19:37 localhost xinetd[1987]: Error parsing attribute server - DISABLING SERVICE [file=/etc/xinetd.d/saned] [line=14]
Apr 18 04:19:37 localhost xinetd[1987]: service sane-port deactivated
Apr 18 04:19:37 localhost xinetd[1987]: sane-port: svc_release with 0 count
Apr 18 04:19:37 localhost synaptic: [RPM] saned-1.0.21-1pclos2010 removed
Apr 18 04:19:37 localhost synaptic: [RPM] sane-backends-iscan-1.0.21-1pclos2010 removed
Apr 18 04:20:49 localhost synaptic: [RPM] libsane1-1.0.21-4pclos2011 installed
Apr 18 04:21:09 localhost synaptic: [RPM] sane-backends-1.0.21-4pclos2011 installed
Apr 18 04:21:51 localhost synaptic: [RPM] saned-1.0.21-4pclos2011 installed
Apr 18 04:27:12 localhost synaptic: [RPM] libsane-hpaio1-3.11.1-2pclos2011 installed
Apr 18 04:27:43 localhost synaptic: [RPM] libsane-hpaio1-3.10.9-2pclos2010 removed
Apr 18 04:29:25 localhost synaptic: [RPM] xsane-0.998-1pclos2011 installed
Apr 18 04:29:27 localhost synaptic: [RPM] xsane-gimp-0.998-1pclos2011 installed
Apr 18 04:32:53 localhost synaptic: [RPM] xsane-0.996-1pclos2010 removed

I think the April 18 entries are the "bring it uptodate" after a bare metal re-install, thanks to seagates fireware update cd.

But /usr/sbin/saned is executable (at least now)
[root@coyote ~]# ls -l /usr/sbin/saned
-rwxr-xr-x 1 root root 61432 Mar  4 15:54 /usr/sbin/saned*

Train of thought here, that MFP is at usb 1-4.  Can I assume that the other devices in this MFP also share this common access point in a usb tree that probably has in excess of 20 branches?

In that event, grepping the messages log for 'usb 1-4' begins to spit out some /interesting/ details:

Apr 27 23:17:45 coyote klogd: usb 1-4: New USB device found, idVendor=04b8, idProduct=0856
Apr 27 23:17:45 coyote klogd: usb 1-4: New USB device strings: Mfr=1, Product=2, SerialNumber=3
Apr 27 23:17:45 coyote klogd: usb 1-4: Product: USB2.0 MFP(Hi-Speed)
Apr 27 23:17:45 coyote klogd: usb 1-4: Manufacturer: EPSON
Apr 27 23:17:45 coyote klogd: usb 1-4: SerialNumber: 4C454C593037313524
Apr 28 01:46:01 coyote klogd: usb 1-4: usbfs: interface 2 claimed by usb-storage while 'usb' sets config #1
Apr 29 01:23:44 coyote klogd: usb 1-4: usbfs: interface 2 claimed by usb-storage while 'usb' sets config #1
Apr 30 02:11:11 coyote klogd: usb 1-4: usbfs: interface 2 claimed by usb-storage while 'usb' sets config #1
Apr 30 14:16:04 coyote klogd: usb 1-4: usbfs: interface 2 claimed by usb-storage while 'usb' sets config #1
Apr 30 14:16:48 coyote klogd: usb 1-4: usbfs: interface 2 claimed by usb-storage while 'usb' sets config #1
Apr 30 14:18:02 coyote klogd: usb 1-4: usbfs: interface 2 claimed by usb-storage while 'usb' sets config #1
Apr 30 14:20:04 coyote klogd: usb 1-4: usbfs: interface 2 claimed by usb-storage while 'usb' sets config #1
Apr 30 14:31:17 coyote klogd: usb 1-4: usbfs: interface 2 claimed by usb-storage while 'usb' sets config #1
May  1 01:57:35 coyote klogd: usb 1-4: usbfs: interface 2 claimed by usb-storage while 'usb' sets config #1
May  1 05:57:03 coyote klogd: usb 1-4: usbfs: interface 2 claimed by usb-storage while 'usb' sets config #1
May  1 05:58:20 coyote klogd: usb 1-4: usbfs: interface 2 claimed by usb-storage while 'usb' sets config #1
May  1 10:20:09 coyote klogd: usb 1-4: usbfs: interface 2 claimed by usb-storage while 'usb' sets config #1
May  1 12:20:58 coyote klogd: usb 1-4: usbfs: interface 2 claimed by usb-storage while 'usb' sets config #1

Is this a clue?

Information overload!

Thanks Galen & Cheers, I gotta go take some sugar pills & find some breakfast.  Yeah, I'm a diabetic old (76) fart, an almost retired tv Chief Engineer.