[sane-devel] Error: Canon Pixma MG7550 hangs

Martin martin.gu at intergga.ch
Tue Feb 24 22:09:18 UTC 2015 

Troels

I don't have access to another computer. But I have installed OpenSUSE 13.2 
and the result is the same as on 13.1.
I did not (yet) test with short delays between the commands, but I compared 
the communication of the Canon software scangear on Windows with the one of 
SANE:

At start ScanGear sends the following (linebreaks added by me):
<?xml version="1.0" encoding="utf-8" ?>
<cmd xmlns:ivec="http://www.canon.com/ns/cmd/2008/07/common/" 
xmlns:vcn="http://www.canon.com/ns/cmd/2008/07/canon/">
  <ivec:contents><ivec:operation>VendorCmd</ivec:operation>
    <ivec:param_set servicetype="scan">
      <ivec:jobID> </ivec:jobID>
      <vcn:ijoperation>ModeShift</vcn:ijoperation>
      <vcn:ijmode>1</vcn:ijmode>
    </ivec:param_set>
  </ivec:contents>
</cmd>

response:
<?xml version="1.0" encoding="utf-8" ?>
<cmd xmlns:ivec="http://www.canon.com/ns/cmd/2008/07/common/" 
xmlns:vcn="http://www.canon.com/ns/cmd/2008/07/canon/">
  <ivec:contents>
    <ivec:operation>VendorCmdResponse</ivec:operation>
    <ivec:param_set servicetype="scan">
      <ivec:response>OK</ivec:response>
      <ivec:response_detail/>
      <vcn:ijoperation>ModeShiftResponse</vcn:ijoperation>
      <vcn:ijresponse>OK</vcn:ijresponse>
      <vcn:ijresponse_detail/>
    </ivec:param_set>
  </ivec:contents>
</cmd>

ScanGear starts a scan with the following command sequece:
<?xml version="1.0" encoding="utf-8" ?>
<cmd xmlns:ivec="http://www.canon.com/ns/cmd/2008/07/common/">
  <ivec:contents>
    <ivec:operation>StartJob</ivec:operation>
    <ivec:param_set servicetype="scan">
      <ivec:jobID>00000001</ivec:jobID>
      <ivec:bidi>1</ivec:bidi>
    </ivec:param_set>
  </ivec:contents>
</cmd>

response:
<?xml version="1.0" encoding="utf-8" ?>
<cmd xmlns:ivec="http://www.canon.com/ns/cmd/2008/07/common/">
  <ivec:contents>
    <ivec:operation>StartJobResponse</ivec:operation>
    <ivec:param_set servicetype="scan">
      <ivec:response>OK</ivec:response>
      <ivec:response_detail/>
      <ivec:jobID>00000001</ivec:jobID>
    </ivec:param_set>
  </ivec:contents>
</cmd>

command:
<?xml version="1.0" encoding="utf-8" ?>
<cmd xmlns:ivec="http://www.canon.com/ns/cmd/2008/07/common/" 
xmlns:vcn="http://www.canon.com/ns/cmd/2008/07/canon/">
  <ivec:contents>
    <ivec:operation>VendorCmd</ivec:operation>
    <ivec:param_set servicetype="scan">
    <ivec:jobID>00000001</ivec:jobID>
    <vcn:ijoperation>ModeShift</vcn:ijoperation>
    <vcn:ijmode>1</vcn:ijmode>
    </ivec:param_set>
  </ivec:contents>
</cmd>

response: some data

command:
<?xml version="1.0" encoding="utf-8" ?>
<cmd xmlns:ivec="http://www.canon.com/ns/cmd/2008/07/common/">
  <ivec:contents>
    <ivec:operation>EndJob</ivec:operation>
    <ivec:param_set servicetype="scan">
      <ivec:jobID>00000001</ivec:jobID>
    </ivec:param_set>
  </ivec:contents>
</cmd>

response:
<?xml version="1.0" encoding="utf-8" ?>
<cmd xmlns:ivec="http://www.canon.com/ns/cmd/2008/07/common/">
  <ivec:contents>
    <ivec:operation>EndJobResponse</ivec:operation>
    <ivec:param_set servicetype="scan">
      <ivec:response>OK</ivec:response>
      <ivec:response_detail/>
      <ivec:jobID>00000001</ivec:jobID>
    </ivec:param_set>
  </ivec:contents>
</cmd>

SANE does not send the VendorCmd. Do you think incorporating the missing 
command sequence could solve the issue? I do not want to damage my MG7550 with 
wrong commands, that's why I am reluctant to experiment with code-changes.

Many thanks for your help
Martin

Am Sonntag, 11. Januar 2015, 14.52:06 schrieb Troels Thomsen:
> Martin,
> ad 1)
> I'm sorry I have no understanding of the protocol what so ever, and have
> spent max ½ hour in the code, to see what the patches I was sent, did.
> ad 2)
> I agree , this IS strange ...
> ad 3)
> I agree , this IS strange ...
> ad 4)
> Frankly, you have so many strange phenomenons going on, so ... you could
> try....
> (not that this a satisfactory solution at all...)
> 
> At this point, I think I would try to test it on another computer, to see
> if all these really strange phenomenons (e.g. 'enabling Ethernet on the
> device, makes usb scanning work' ) somehow is related to hardware or the
> Linux installation. Just to make sure....
> 
> Sorry, its not much help.
> 
> /Troels
> 
> 2015-01-07 23:26 GMT+01:00 Martin <martin.gu at intergga.ch>:
> > Troels
> > 
> > I agree with you, we have to be careful on the test methodology.
> > My knowledge in c is limited since I wrote my last c-program about 20
> > years
> > ago. But I could try.
> > But before trying to modify the source code I have some questions:
> > 
> > 1) Did you analyze the USB-calls of the Canon-software on windows and
> > compare
> > them to the SANE-sequence? What is the meaning of the first sequece 55 53
> > 42
> > 43 F0 36 04 0B...?
> > 
> > 2) What is the difference between scanimage and xsane? Is it only the
> > timing?
> > 
> > 3) What is the difference between xsane and the gimp plugin? Is it only
> > the
> > timing? Why does the gimp plugin only work after xsane?
> > 
> > 4) Would it be worth while to make a few tests with the printer mode, i.e.
> > to
> > try to start xsane after printing?
> > 
> > Martin
> > 
> > > > If the scanner does not work on a USB port it is quite unlikely to
> > > > work
> > > > over the network: the bjnp network code just carries the USB exchange
> > > > over IP.
> > > 
> > > I mostly agree, though Rolf asks specifically for the USB trace, because
> > > (as I understood) it differs slightly from the IP traces I had already
> > > submitted.
> > > (fix usb first , then patch it for IP, if need be)
> > > 
> > > One advice:
> > > Be very paranoid on your test methodology. I looked stupid a couple of
> > > times (sorry Rolf!), because I was fooled by the USB interface and/or
> > > the
> > > Canon firmware.
> > > When testing a new patch, I tried some settings in random (DPI ,
> > > colour/black-n-white etc) and reported the success/failures.
> > > BUT the thing is, that the Canon MF8230 firmware can go into bad-mode,
> > 
> > when
> > 
> > > it has received some strange commands, and the following scans might
> > > work
> > > slightly strange for that reason. I was even fooled by the phenomenon,
> > 
> > that
> > 
> > > scanning a 1 by 1 cm square works, but scanning the whole A4 didn't.
> > 
> > (Well
> > 
> > > ... there WERE some differences between what worked through USB and IP
> > > at
> > > this point in time. I'm guessing it was something timing-critical, but I
> > > never dissected that last patch from Rolf which fixed it)
> > > 
> > > So I started rebooting the scanner whenever I was in doubt.
> > > (disconnect/connect the USB interface plus move Cannon unit from to
> > > "printer mode" and back to "scanner mode" was normally enough)
> > > 
> > > 
> > > So what do we do? / What is status?
> > > 
> > > Since scanning CAN work , one could argue the back-end DOES send all the
> > > necessary+correct commands to the printer.
> > > If timing is the critical parameter (backed by the fact that something
> > > as
> > > stupid as enabling IP on the printer-side, suddenly makes it work now
> > > and
> > > then), maybe we should try to insert an ugly time-delay before every USB
> > > write? If that changes anything to the better, we can start to hunt down
> > > where it is really necessary, and hopefully find a nicer approach
> > > afterwards?
> > > 
> > > Martin, are you comfortable with pulling the code out of the GIT
> > > repository, compile , install and experiment with inserting delays?
> > > I think there is something called "SANE interface USB" sanei_usb.c/h.
> > 
> > That
> > 
> > > was maybe a place to start.
> > > 
> > > Louis, what is the odds of success for this approach?
> > > 
> > > :-)
> > > 
> > > /Troels
> > > 
> > > 2015-01-03 20:43 GMT+01:00 Martin <martin.gu at intergga.ch>:
> > > > Am Samstag, 3. Januar 2015, 11.16:10 schrieb Louis Lagendijk:
> > > > > On Fri, 2015-01-02 at 22:59 +0100, Martin wrote:
> > > > > > Hi Troels
> > > > > > 
> > > > > > > *Is it USB3 (on the computer)?
> > > > > > 
> > > > > > no, it is USB2
> > > > > > 
> > > > > > > *Do you have access to another computer, where you can try it -
> > 
> > Does
> > 
> > > > it
> > > > 
> > > > > > > work excactly the same strange way ... ?
> > > > > > 
> > > > > > No,I don't have access to another computer. But I have also
> > 
> > Windows 7
> > 
> > > > > > installed. With Windows 7 and the Canon-software it works OK.
> > > > > > 
> > > > > > > *Does unit have ethernet connection / can you try that way?
> > > > > > 
> > > > > > Yes. With the ethernet connection the printer works. Thus the
> > > > 
> > > > IP-address
> > > > 
> > > > > > below is correct.
> > > > > > But I cannot access the scanner.
> > > > > > 
> > > > > >     xsane pixma:MG7500_192.168.0.3
> > > > > > 
> > > > > > results in the error message "illegal argument". Same thing with
> > > > > > 
> > > > > >     scanimage --help -d bnjp://192.168.0.3:8612
> > > > > 
> > > > > Hi Martin
> > > > > If the scanner does not work on a USB port it is quite unlikely to
> > 
> > work
> > 
> > > > > over the network: the bjnp network code just carries the USB
> > > > > exchange
> > > > > over IP.
> > > > > Anyhow, you can find the name with scanimage -L (but make sure that
> > 
> > port
> > 
> > > > > 8612 is not blocked by a firewall (see the man-page for sane-pixma
> > > > > in
> > > > > the latest sources from GIT. you need to enable port 8612 for both
> > > > > incoming and outgoing traffic.
> > > > > 
> > > > > Louis
> > > > 
> > > > Hi Louis
> > > > 
> > > > The necessary ports on the firewall are open, yet scanimage -L does
> > > > not
> > > > find
> > > > the scanner.
> > > > 
> > > > But this lead to another discovery:
> > > > I enabled LAN on the MG7550. And although the unit is connected via
> > 
> > USB 2
> > 
> > > > xsane sometimes works?! But with LAN disabled xsane never works
> > > > without
> > > > prior
> > > > scanimage -T. And once xsane or scanimage -T failed all succeeding
> > 
> > calls
> > 
> > > > fail
> > > > until the scanner is switched off and on again.
> > > > 
> > > > Log of xsane working:
> > > > [sanei_debug] Setting debug level of pixma to 21.
> > > > [pixma] pixma is compiled with pthread support.
> > > > [pixma] pixma version 0.17.13
> > > > [pixma] pixma_collect_devices() found Canon PIXMA MG7500 Series at
> > > > libusb:001:007
> > > > [pixma] pixma_find_scanners() found 1 devices
> > > > [pixma] pixma_collect_devices() found Canon PIXMA MG7500 Series at
> > > > libusb:001:007
> > > > [pixma] pixma_open(): Canon PIXMA MG7500 Series
> > > > [pixma] *mp150_open***** This is a generation 4 scanner.  *****
> > > > [pixma] [pixma] Reader task id=139772037314304 (threaded)
> > > > Reader task started
> > > > [pixma]
> > > > [pixma] pixma_scan(): start
> > > > [pixma]   line_size=4605 image_size=4895115 channels=3 depth=8
> > > > [pixma]   dpi=300x300 offset=(0,0) dimension=1535x1063
> > > > [pixma]   gamma_table=0x1f201e0 source=0
> > > > [pixma]   threshold=127 threshold_curve=0
> > > > [pixma]   ADF page count: 0
> > > > [pixma] OUT  T=58.770 len=288
> > > > [pixma]  00000000:3c 3f 78 6d 6c 20 76 65  72 73 69 6f 6e 3d 22 31
> > > > [pixma]  00000010:2e 30 22 20 65 6e 63 6f  64 69 6e 67 3d 22 75 74
> > > > ...
> > > > Log of xsane not working:
> > > > pixma] pixma_collect_devices() found Canon PIXMA MG7500 Series at
> > > > libusb:001:008
> > > > [pixma] pixma_find_scanners() found 1 devices
> > > > [pixma] pixma_collect_devices() found Canon PIXMA MG7500 Series at
> > > > libusb:001:008
> > > > [pixma] pixma_open(): Canon PIXMA MG7500 Series
> > > > [pixma] *mp150_open***** This is a generation 4 scanner.  *****
> > > > [pixma] INTR T=1.249 len=32
> > > > [pixma]  00000000:00 00 00 01 00 00 00 00  00 00 00 00 00 00 00 00
> > > > [pixma]  00000010:00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00
> > > > [pixma]
> > > > [pixma] Sending time: '15/01/03 15:32'
> > > > [pixma] OUT  T=1.250 len=36
> > > > [pixma]  00000000:eb 80 00 00 00 00 00 00  00 00 00 00 00 00 00 14
> > > > [pixma]  00000010:31 35 2f 30 31 2f 30 33  20 31 35 3a 33 32 00 00
> > > > [pixma]  00000020:00 00 00 53
> > > > [pixma]
> > > > [pixma] IN   T=21.251 len=-9
> > > > [pixma]   ERROR: ETIMEDOUT
> > > > [pixma]
> > > > ...
> > > > 
> > > > Martin
> > > > 
> > > > 
> > > > --
> > > > sane-devel mailing list: sane-devel at lists.alioth.debian.org
> > > > http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/sane-devel
> > > > Unsubscribe: Send mail with subject "unsubscribe your_password"
> > > > 
> > > >              to sane-devel-request at lists.alioth.debian.org
> > 
> > --
> > sane-devel mailing list: sane-devel at lists.alioth.debian.org
> > http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/sane-devel
> > Unsubscribe: Send mail with subject "unsubscribe your_password"
> > 
> >              to sane-devel-request at lists.alioth.debian.org

More information about the sane-devel mailing list