http://bugzilla.opensuse.org/show_bug.cgi?id=1052793 http://bugzilla.opensuse.org/show_bug.cgi?id=1052793#c1 Antonio Larrosa <alarrosa@suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |alarrosa@suse.com --- Comment #1 from Antonio Larrosa <alarrosa@suse.com> --- I think this shouldn't fix it, but just to discard some possible causes of problems... can you try running okular or kate with the following command and see if that changes anything? QT_DISABLE_PRINTER_DISCOVERY=1 okular -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1052793 http://bugzilla.opensuse.org/show_bug.cgi?id=1052793#c3 Fabian Vogt <fvogt@suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |fvogt@suse.com Flags| |needinfo?(alarrosa@suse.com | |) --- Comment #3 from Fabian Vogt <fvogt@suse.com> --- (In reply to Eric McAlvin from comment #0)

Here is what is printed on the console with QT logging enabled: $ QT_LOGGING_RULES=org.kde.okular.core=true okular org.kde.okular.core: Executing "lpr" with arguments ("-P", "", "-#1", "-J", "Dell.pdf", "-r", "/tmp/okular_S21220.ps")

That looks obvious: "-P", "" means okular didn't specify a printer at all for some reason. That means QPrinter::printerName() returns an empty value, which means either the backend or the printer selection dialog is broken. I don't see any upstream changes in either of those, so I still suspect that antlarr's patch might play a role here. @Antonio: Is that likely? If so, can you provide a package without that patch? -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1052793 http://bugzilla.opensuse.org/show_bug.cgi?id=1052793#c5 Fabian Vogt <fabian@ritter-vogt.de> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |fabian@ritter-vogt.de --- Comment #5 from Fabian Vogt <fabian@ritter-vogt.de> --- (In reply to Antonio Larrosa from comment #4)

@Eric, could you try running this:

[...]

That's basically a rebuild of the same packages you should already have installed. So we're just finding out if this is related to Mageia-2016-0055 or not (note I don't think so, since we don't include those patches on Leap).

I don't think so either, as the mageia bug looks totally unrelated. In fact, I think it's very much unrelated to any part of the printer subsystem. Okular simply forwards the returned QPrinter from the QPrintDialog to the lpr command line, at which point the printer queue's name is empty. So either the print dialog is broken (unlikely) or the source data is wrong.

@Fabian, I'll try to generate qt packages without the remote printer discovery patch, but note that setting QT_DISABLE_PRINTER_DISCOVERY=1 should use the same code as if the patch was disabled.

Yes, it should indeed, but I simply don't see any other change that could cause this behaviour... -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1052793 http://bugzilla.opensuse.org/show_bug.cgi?id=1052793#c7 --- Comment #7 from Eric McAlvin <ericj@mcalvin.com> --- I have installed the packages you listed, ran rccups restart, and also restarted my window manager (ctrl-alt-backspace/backspace) as zypper ps -s showed open files. However, the symptoms are still the same. $ QT_LOGGING_RULES=org.kde.okular.core=true okular org.kde.okular.core: "/usr/lib64/qt5/plugins/okular/generators/okularGenerator_poppler.so" org.kde.okular.core: Output DPI: QSizeF(94.1467, 94.0741) org.kde.okular.core: request observer=0x1d91a00 692x894@0 org.kde.okular.core: request observer=0x1d91a00 692x894@1 org.kde.okular.core: request observer=0x1d91a00 692x894@2 org.kde.okular.core: sending request observer=0x1d91a00 692x894@0 async == true isTile == false org.kde.okular.core: request observer=0x1d91a00 692x894@0 org.kde.okular.core: request observer=0x1d91a00 692x894@1 org.kde.okular.core: request observer=0x1d91a00 692x894@2 org.kde.okular.core: request observer=0x1d91a00 672x868@0 org.kde.okular.core: request observer=0x1d91a00 672x868@1 org.kde.okular.core: request observer=0x1d91a00 672x868@2 org.kde.okular.core: sending request observer=0x1d91a00 672x868@0 async == true isTile == false org.kde.okular.core: request observer=0x1d91a00 672x868@0 org.kde.okular.core: request observer=0x1d91a00 672x868@1 org.kde.okular.core: request observer=0x1d91a00 672x868@2 org.kde.okular.core: sending request observer=0x1d91a00 672x868@1 async == true isTile == false org.kde.okular.core: request observer=0x1d91a00 672x868@2 org.kde.okular.core: sending request observer=0x1d91a00 672x868@2 async == true isTile == false org.kde.okular.core: request observer=0x1f7ec40 440x569@0 org.kde.okular.core: request observer=0x1f7ec40 440x569@1 org.kde.okular.core: request observer=0x1d91a00 672x868@2 org.kde.okular.core: sending request observer=0x1f7ec40 440x569@0 async == true isTile == false org.kde.okular.core: sending request observer=0x1f7ec40 440x569@1 async == true isTile == false org.kde.okular.core: Executing "lpr" with arguments ("-P", "", "-#1", "-J", "Dell Client Statement on Intel AMT Advisory (INTEL-SA-00075).pdf", "-r", "/tmp/okular_M22698.ps") lpr: The printer or class does not exist. ^C One thing I did notice is that when you bring the print dialog window up, it defaults to print to file (PDF) rather than my default printer. I don't think 42.1 (or 42.2) behaved this way. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1052793 http://bugzilla.opensuse.org/show_bug.cgi?id=1052793#c9 --- Comment #9 from Eric McAlvin <ericj@mcalvin.com> --- Created attachment 736340 --> http://bugzilla.opensuse.org/attachment.cgi?id=736340&action=edit okular stdout/stderr I have installed your RPM and re-ran okular. Output is attached. Note that my printer names have been changed to printer0, printer1, to protect the super secret names of my printers :). Note that the problem does still indeed exist with this new package version. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1052793 http://bugzilla.opensuse.org/show_bug.cgi?id=1052793#c11 Antonio Larrosa <alarrosa@suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |CONFIRMED --- Comment #11 from Antonio Larrosa <alarrosa@suse.com> --- Sorry for taking a few days to answer. The logs are quite interesting and show that indeed there's some problem in Qt when retrieving the printer information. Btw, you have quite a long list of printers. Are all those printers defined in a print server? (or several print servers) or are they configured locally? Btw, yes, it's ok to revert the previous sets of packages. Since the way printer information is obtained depends on the printer being remote or local, I included again the "remotePrinterDiscovery" patch that I removed in the previous test package and added some more debug output on that particular point. Can you install this package and try again? zypper in http://download.opensuse.org/repositories/home:/alarrosa:/branches:/openSUSE... Btw, it would be great if you can test twice and send both logs. First running okular as usual and then running okular with "QT_DISABLE_PRINTER_DISCOVERY=1 okular", as both will use different ways to get printer information and should show different output. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1052793 http://bugzilla.opensuse.org/show_bug.cgi?id=1052793#c13 --- Comment #13 from Eric McAlvin <ericj@mcalvin.com> --- Created attachment 736865 --> http://bugzilla.opensuse.org/attachment.cgi?id=736865&action=edit okular with special build of libQt5PrintSupport5-5.6.2-15.1.x86_64 -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1052793 http://bugzilla.opensuse.org/show_bug.cgi?id=1052793#c15 --- Comment #15 from Eric McAlvin <ericj@mcalvin.com> --- (In reply to Eric McAlvin from comment #12)

All of these printers are defined on the cups print server; none are local.

I have tested okular with the new package. The issue still exists. But I have attached both output attempts.

Sorry I forgot to mention - they are all defined on a pair of print servers. So, we have a round robin DNS record called 'printserver' and both are cups servers with the same 149 printers defined. My cups client points to that 'printserver' name. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1052793 http://bugzilla.opensuse.org/show_bug.cgi?id=1052793#c17 --- Comment #17 from Andreas Seeg <andreas.seeg@posteo.de> --- We seem to have hit the same problem with 42.3: Printing from KDE applications fails silently. We have changed our configuration (two cups servers in cups-browsed.conf) to a single server in cups-browsed.conf, and then to no entry in cups-browsed.conf but a single entry in client.conf, and the problem still persists. We do not employ a round robin dns. Printing from Evince or other non-KDE applications works fine. Printing with an old 42.1 system via the same print servers using KDE applications works fine. We did "zypper in http://download.opensuse.org/repositories/home:/alarrosa:/branches:/openSUSE..." and I have attached the output of two application (okular and kate) after this change (okular and kate output before/after libQt5PrintSupport5-5.6.2-8.1.x86_64.rpm). I chose to keep a single server in cups-browsed.conf and no entry in client.conf and restarted cups and cups-browsed after installing the package. Printing to PDF and then using Evince to print the resulting PDF can be used as a workaround. We did not have a local printer to test if the problem expands to local printing, too, so this problem might only affect printing via remote CUPS servers. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1052793 http://bugzilla.opensuse.org/show_bug.cgi?id=1052793#c19 --- Comment #19 from Andreas Seeg <andreas.seeg@posteo.de> --- While printing is still broken for the KDE apps we use, we noticed that Kile can print just fine despite being a Qt application. We re-checked Okular and Kate for 42.3 just to be sure, and their printing jobs still silently fail. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1052793 http://bugzilla.opensuse.org/show_bug.cgi?id=1052793#c21 Henry Laurent <laurent.henry@ehess.fr> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |laurent.henry@ehess.fr --- Comment #21 from Henry Laurent <laurent.henry@ehess.fr> --- Hello, i am encoutering the same issue just recently. I don't know when the issue begun exactly but things were fine lately. I don't have anything new to add but i can help to troubleshoot if necessary. I am just notincing okular propose a list of printers from network automatic browsing not from my local yast/cups config. The local config works fine with every other app, like libre office. Regards -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1052793 http://bugzilla.opensuse.org/show_bug.cgi?id=1052793#c23 --- Comment #23 from Michael Lashkevich <lashkevi@landau.ac.ru> --- In addition, qt5 applications mix up some queues: I send a job to a queue, and it goes to another queue. -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1052793 http://bugzilla.opensuse.org/show_bug.cgi?id=1052793#c25 --- Comment #25 from Eric McAlvin <ericj@mcalvin.com> --- (In reply to Antonio Larrosa from comment #24)

Hmmm, anyone having the issue can check if the same problem happens on a Leap 15 system?

I just tested printing from Okular on openSUSE 15.0 and it worked this time. My setup is the same as when I initially opened this bugzilla (CUPS with a 100+ print queues). I am just running a fresh install of Leap 15.0 and it worked! -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.opensuse.org/show_bug.cgi?id=1052793 http://bugzilla.opensuse.org/show_bug.cgi?id=1052793#c27 --- Comment #27 from Eric McAlvin <ericj@mcalvin.com> --- (In reply to Antonio Larrosa from comment #26)

(In reply to Eric McAlvin from comment #25)

...

(In reply to Antonio Larrosa from comment #24)

...

Hmmm, anyone having the issue can check if the same problem happens on a Leap 15 system?

I just tested printing from Okular on openSUSE 15.0 and it worked this time. My setup is the same as when I initially opened this bugzilla (CUPS with a 100+ print queues). I am just running a fresh install of Leap 15.0 and it worked!

Thanks for the quick reply, Eric. This means the problem was probably fixed between Qt 5.6.x and 5.9.x, so we have some hint on the solution for Leap 42.3 .

Btw, just to be sure I understood you. When you say that the setup is the same as when you opened the bugzilla entry, you mean you tested with the round robin DNS record?

Can anyone else having the issue also try Leap 15.0 and see if that works for their cases as well?

That is correct - the backend infrastructure has not changed (still a round robin record, backend CUPSD still the same version/same OS). The only change is my workstation was rebuilt as Leap 15.0. -- You are receiving this mail because: You are on the CC list for the bug.