Garbage bitmaps printed on left margin in ubuntu testpage on A4 on Samsung printers

Same here with the Samsund ML-1610.

Upon further inspection I found that it is indeed related to the paper size, when I select letter the garbage is gone.

There is a slightly deterministic pattern in the prints, in two prints from the same document the garbage is about 50% the same.

Printing in opera(browser) is completely non-readable using the new driver, the graphics look ok, but the text is a very light gray instead of black. Printing the sampe page (this one) in firefox had the garbage band, but is otherwise readable so it might be something to do with the way opera generates the document to print.

I now use the old driver as workaround. There selecting a wrong Papersize (e.g A5 instead of A4) has a simular result - garbidge.

Same here w/Samsung ML-1610.

Also, the image quality is very low when compared to the foomatic/gdi driver.

The bug is easily reproducible, and since the foomatic/gdi driver works, it's probaby caused by the splix package.

Can you go to the following two pages

http://openprinting.org/show_driver.cgi?driver=splix
http://openprinting.org/show_driver.cgi?driver=splix2

and download the DEB packages of the two versions of the SpliX driver from there? Take care to download the packages for your system architecture and LSB 3.2.

Then install at first the driver version 2.0.0 rc2 (with "sudo dpkg -i ..."), remove your print queue(s) for your Samsung printer(s) and recreate them, selecting the driver entry in system-config-printer with "OpenPrinting". Now try again to print. Does the problem go away?

If not, try the same thing with version 1.1.1. Does this work?

The OpenPrinting splix packages (both 1.1.1 and 2.0.0rc2) don't work at all.

CUPS fails with an error saying "can't write Page 1 header"

Same error here.

Can Berk Güder wrote:
> The OpenPrinting splix packages (both 1.1.1 and 2.0.0rc2) don't work at
> all.
>
> CUPS fails with an error saying "can't write Page 1 header"
>
>

Can Berk Güder, elleP, can you send me a full error_log of your printing attampts with the OpenPrinting driver packages? Otherwise I cannot find out where the "can't write Page 1 header" comes from.

Another possible cause of the problem can be the pstopdf filter. Can you all replace your /usr/lib/cups/filter/pstopdf by the attached file? Please do not forget to make it executable

sudo chmod 755 /usr/lib/cups/filter/pstopdf

Can you print now?

Here's my error log. I also had trouble with libpam-smbpass today, so there might be some noise.

Oh, and I'm using OpenPrinting Splix 2.0.0rc2 in Jobs 56 & 57, and OpenPrinting Splix 1.1.1 in Job 58.

The "filter not available" errors are due to me removing the openprinting packages, I believe.

And the error log seems to point to the pdftoraster filter.

This is the print log (job 508) using the OpenPrinting driver.
This uses the replacement pstopdf filter, but unfortunately it still fails.

Till Kamppeter wrote:
> Can Berk Güder, elleP, can you send me a full error_log of your printing
> attampts with the OpenPrinting driver packages? Otherwise I cannot find
> out where the "can't write Page 1 header" comes from.
>
> Another possible cause of the problem can be the pstopdf filter. Can you
> all replace your /usr/lib/cups/filter/pstopdf by the attached file?
> Please do not forget to make it executable
>
> sudo chmod 755 /usr/lib/cups/filter/pstopdf
>
> Can you print now?
>
> ** Attachment added: "Replacement pstopdf filter"
> http://launchpadlibrarian.net/19398553/pstopdf
>
> ** Changed in: splix (Ubuntu)
> Status: Confirmed => Incomplete
>
>

I have fixed another bug. Please test with this new version of pstopdf, not with the one of the previous comment.

Can you please try the new replacement pstopdf with the SpliX driver from Ubuntu. So we can check at first whether the pstopdf fixes the bug originally reported here.

Nope, it doesn't solve the problem. The garbage is still there.

I can confirm this problem with my ML-1610. I haven't tested the new filter yet.

FYI, this happens if I print a PDF from evince. If I print it with "lp foo.pdf", the output is fine, so that's a temporary workaround.

I confirm that the new script fixes the "%% No page comment in header" (or similar) error message, but not the garbage. I tested with both the splix and foomatic/gdi driver.

I tried:

 - use system-config-printer to create a new one; I used AppSocket/JetDirect localhost:9100, but it shouldn't really matter
 - select Samsung ML-1610; either driver is fine, the problem happens with both splix and foomatic
 - Open any PDF document with evince, I used

     evince /usr/share/doc/shared-mime-info/shared-mime-info-spec.pdf

   and printed page 17 into a PS file (/tmp/out.ps).

 - cupsfilter -m application/vnd.cups-pdf -p /etc/cups/ppd/test.ppd /tmp/out.ps > /tmp/out.pdf

/tmp/out.pdf doesn't have garbage. The conversion produced error messages

ERROR: No %%BoundingBox: comment in header!
ERROR: No %%Pages: comment in header!

These error messages are fixed if I use the new pstopdf filter.

So this doesn't reproduce the garbage :(

This is the result with the new pstopdf filter from above test.

Can you try this pstopdf?

Can Berk Güder, for your problems with the OpenPrinting packages of SpliX, can you report a bug on

http://bugs.linux-foundation.org/

product "OpenPrinting" and component "Driver packages".

Can Berk Güder, please attach also your complete error logs there.

Till, your new pstopdf from comment 21 works perfectly!

I confirm that changing "-dHaveTransparency=false" to "-dHaveTransparency=true" works, too.

I also verified that dropping "-dHaveTransparency=false" works. So adding "-dPDFSETTINGS=/printer" is sufficient.

The attached pstopdf fixes this bug according to pitti.

This bug was fixed in the package cups - 1.3.9-3

---------------
cups (1.3.9-3) experimental; urgency=low

  [ Till Kamppeter ]
  * debian/filters/pstopdf: Fixed several bugs in the pstopdf filter. First,
    removed the use of CUPS' pstops filter for inserting option settings. This
    also inserts PJL headers and then Ghostscript cannot convert the PostScript
    to PDF in the next step. Fixed also the sed magic so that the paper size
    and the margins get really read from the PPD and fixed the calculation of
    the top and bottom margins, they were exchanged. Fixes LP: #289759,
    LP: #292690, LP: #282186. Possible fix for LP #293883.

  [ Martin Pitt ]
  * debian/local/apparmor-profile: Allow dnssd backend to create various less
    common network protocols (x25, appletalk, etc.) for detection. Also allow
    it to read /proc/*/net/, which the bonjour avahi library apparently uses.
    (LP: #254022)

 -- Martin Pitt <email address hidden> Wed, 29 Oct 2008 11:41:38 +0100

None of the new pstopdf filters work for me.

Accepted into intrepid-proposed, please test and give feedback here. Please see https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Thank you in advance!

Can Berk Güder, you are probably hitting the problem of Launchpad considering the plain text attachments as HTML.

To get my pstopdf filter correctly, either install the fixed CUPS package from -proposed, as pitti tells in the previous comment or do

rm pstopdf
wget http://launchpadlibrarian.net/19430160/pstopdf
sudo cp pstopdf /usr/lib/cups/filter/pstopdf
sudo chmod 755 /usr/lib/cups/filter/pstopdf

Nope, I used wget. It prints, but the garbage is still there.

Can Berk Güder, pitti, seems that now we are dependent on the input files.

Can everyone of you try several files and applications and post files which produce the garbage. Can you also tell with print situations are garbage-free now?

Can you also test both the SpliX and the gdi drivers (if applicable for your printer model)?

SpliX drivers fail, Foomatic/gdi drivers work.

I've tried:

* Printing a PDF from Acroread
* Printing a web page from Firefox
* Printing a test page from system-config-printer
* Printing a test page from the CUPS web interface

with both drivers.

In each case, the SpliX drivers produced garbage, while the
Foomatic/gdi drivers worked.

pitti, did you test your printer with various different input files? Did you also test the SpliX driver? Can you test the SpliX packages from OpenPrinting, both the 1.1.1 and 2.0.0 versions?

Till Kamppeter [2008-11-06 21:30 -0000]:
> pitti, did you test your printer with various different input files?

With two, plus the test page. It didn't seem to make a difference. As
I said, I used
/usr/share/doc/shared-mime-info/shared-mime-info-spec.pdf, since
that's available in a default install.

> Did you also test the SpliX driver?

No, I'll do that today.

I confirm that when using splix (the packaged one in intrepid), I
still get the garbage.

Probably SpliX has a problem here, too. Adding SpliX task.

pitti, can you test with the OpenPrinting packages of both SpliX versions?

Can you also try the following?

First, download this file:

http://www.openprinting.org/download/printing/ghostscript-cups-patches/pdftoraster.c

This is an alternative Ghostscript-based pdftoraster filter.

Install packages needed to compile it:

sudo apt-get install libcups2-dev libcupsimage2-dev

Compile it:

gcc `cups-config --image --libs` -DBINDIR=/usr/bin -DGS=gs -o pdftoraster pdftoraster.c

Install it:

sudo mv /usr/lib/cups/filter/pdftoraster /usr/lib/cups/filter/pdftoraster.orig
sudo cp pdftoraster /usr/lib/cups/filter/
sudo chmod 755 /usr/lib/cups/filter/pdftoraster

Now try to print again. Does the garbage go away?

It's impossible to compile the file on my machine.

Can you use the following command line instead:

gcc -lcups -lcupsimage -o pdftoraster pdftoraster.c

I tested the openprinting.org packages according to https://bugs.launchpad.net/ubuntu/+source/splix/+bug/291812/comments/3:

With splix 2, printer configuration succeeds, but when printing something from evince, I get:

E [10/Nov/2008:16:52:53 +0100] PID 6701 (/opt/OpenPrinting-SpliX/cups/lib/filter/rastertoqpdl) stopped with status 22!
E [10/Nov/2008:16:52:56 +0100] [Job 377] Can't write page 1 header
E [10/Nov/2008:16:52:56 +0100] PID 6698 (/usr/lib/cups/filter/pdftoraster) stopped with status 1!
E [10/Nov/2008:16:52:56 +0100] [Job 377] Job stopped due to filter errors.

I get the exact same problem when trying to print the test page in s-c-p (which doesn't use pdftops AFAIK).

With splix 1, it looks almost identical:

E [10/Nov/2008:16:59:25 +0100] PID 7044 (/opt/OpenPrinting-SpliX/cups/lib/filter/rastertospl2) stopped with status 22!
E [10/Nov/2008:16:59:27 +0100] [Job 379] Can't write page 1 header
E [10/Nov/2008:16:59:27 +0100] PID 7041 (/usr/lib/cups/filter/pdftoraster) stopped with status 1!
E [10/Nov/2008:16:59:27 +0100] [Job 379] Job stopped due to filter errors.

If I use the intrepid final splix 2.0 and above new pdftoraster.c, I still get garbage.

@Till Kamppeter

Even if I tried your commands, the printer still prints garbage.

PS I rebooted cups

cups copied to intrepid-updates.

Can Berk Güder, elleP, pitti, I have replaced the SpliX packages on OpenPrinting, as there were built with a broken compiler. Now they should work correctly. Please do your tests with these packages again.

https://bugs.edge.launchpad.net/splix/+bug/292690/comments/6

I'm getting the "%Page Comment in header!" error when trying to print from Ubuntu 8.10 on my HP Deskjet 5550. Same issue has been raised at https://answers.launchpad.net/hplip/+question/50975 for HP 3650.

(splix packages look like they don't apply to HP, is that right?)

With the updated splix2 package from openprinting.org I still get the same failure.

Cross-testing: I get the same garbage when directly printing a PDF with "lp" or the test page (which doesn't use pdftops) with the intrepid splix package. If I downgrade the splix package to hardy's version (1.1.1), printing from evince as well as test page works. This sufficiently proves that it is a regression in splix 2, and not a problem in pdftops any more.

Reverting no-crash-on-bad-papersize.patch in the intrepid package doesn't make a difference (as expected, since it only influences the page height, however, the garbage is at the left margin).

I attach the error_log (with debugging) when using the OpenPrinting.org splix 2 package. NB the

  D [13/Nov/2008:09:44:26 +0100] [Job 393] /opt/OpenPrinting-SpliX/cups/lib/filter/rastertoqpdl: No such file or directory

which is bogus, the file exists and is executable.

As asked by Till, this is the debug-ful error_log with openprinting-splix_1.1.1-4lsb3.2_i386.deb.

Please ignore my troubles with the OpenPrinting packages, they occurred because I forcefully ignored the "lsb" dependency (because it pulls in such a lot of stuff I don't need on this desktop box, such as postfix (!), rpm, alien, the entire set of development packages, Qt, etc.

Now the openprinting.org splix 2 package gives exactly the same result, garbage on the left margin.

So this is definitely an upstream regression as the problem is independent of the packaging and does not occur in SpliX 1.1.1. As a workaround users should use the Hardy or the OpenPrinting package of SpliX 1.1.1 if the printer is supported by this version. Also the "gdi" driver can be used for many (older) printers.

Banding is fixed in latest SVN-head at
https://sourceforge.net/svn/?group_id=175815

This bug was fixed in the package splix - 2.0.0~rc2-0ubuntu6

---------------
splix (2.0.0~rc2-0ubuntu6) jaunty; urgency=low

  * debian/patches/all-papersize-fixes-from-svn.patch: Applied all paper-size-
    related fixes from the SVN repository up to November 25 (LP: #292690).
  * debian/patches/no-crash-on-bad-papersize.patch: Removed, merged into
    all-papersize-fixes-from-svn.patch.

 -- Till Kamppeter <email address hidden> Wed, 26 Nov 2008 09:19:45 +0100

Thank you for the hint, elleP, I have applied the fix from SVN and uploaded it to Jaunty. An SRU for Intrepid will come soon.

Accepted splix into intrepid-proposed, please test and give feedback here. Please see https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Thank you in advance!

I built the package locally from the queue and tested it, works fine now. Thanks, Till!

> Accepted splix into intrepid-proposed, please test and give feedback here.

No effect. I installed 2.0.0~rc2-0ubuntu2.2, restart machine, print out text for OpenOffice.org Writer and get garbage.

Kubuntu 8.10, AMD64, Samsung ML-1615.

That is not the correct package, you have to wait for 2.0.0~rc2-0ubuntu2.3, which should arrive on your mirror in somew hours. Try "sudo apt-get update", the file is perhaps already there.

Well. I downloaded 2.0.0~rc2-0ubuntu6 :) Manually. It is the latest (26-11-2008) package at archive.ubuntu.com. I think, it is for jaunty, but it works.

Still waiting for 2.3.

I installed the 2.0.0~rc2-0ubuntu2.3 from proposed, reconfigured my printer and it appears to be fixed.
Thanx all!

This bug was fixed in the package splix - 2.0.0~rc2-0ubuntu2.3

---------------
splix (2.0.0~rc2-0ubuntu2.3) intrepid-proposed; urgency=low

  * debian/patches/all-papersize-fixes-from-svn.patch: Applied all paper-size-
    related fixes from the SVN repository up to November 25 (LP: #292690).
  * debian/patches/no-crash-on-bad-papersize.patch: Removed, merged into
    all-papersize-fixes-from-svn.patch.

 -- Till Kamppeter <email address hidden> Wed, 26 Nov 2008 09:19:45 +0100