I confirm.  I haven't had occasion to use s-c-p lately until yesterday, and I noticed this.  Launching from a terminal as root worked, but launching from MCC failed reliably.

CC: (none) => ftg

Here is the stdout/stderr from running mcc from the command line:

[ftg@ftgme2 ~]$ mcc
LOG **: NP_Initialize at /usr/lib/libDrakX/mygtk2.pm line 589.
LOG **: NP_Initialize succeeded at /usr/lib/libDrakX/mygtk2.pm line 589.
LOG **: NP_Initialize at /usr/lib/libDrakX/mygtk2.pm line 589.
LOG **: NP_Initialize succeeded at /usr/lib/libDrakX/mygtk2.pm line 589.
LOG **: NP_Initialize at /usr/lib/libDrakX/mygtk2.pm line 589.
LOG **: NP_Initialize succeeded at /usr/lib/libDrakX/mygtk2.pm line 589.
LOG **: NP_Initialize at /usr/lib/libDrakX/mygtk2.pm line 589.
LOG **: NP_Initialize succeeded at /usr/lib/libDrakX/mygtk2.pm line 589.
"/usr/sbin/drakmenustyle" is not executable [Menus] at /usr/libexec/drakconf line 823.
"/usr/sbin/drakbackup" is not executable [Backups] at /usr/libexec/drakconf line 823.
"/usr/sbin/tomoyo-gui" is not executable [Tomoyo Policy] at /usr/libexec/drakconf line 823.
java version "1.7.0_45"
OpenJDK Runtime Environment (mageia-2.4.3.2.mga4-x86_64 u45-b15)
OpenJDK 64-Bit Server VM (build 24.45-b08, mixed mode)
Traceback (most recent call last):
  File "/usr/share/system-config-printer/system-config-printer.py", line 2136, in <module>
    main(show_jobs)
  File "/usr/share/system-config-printer/system-config-printer.py", line 2108, in main
    mainwindow = GUI()
  File "/usr/share/system-config-printer/system-config-printer.py", line 203, in __init__
    PlugWindow = gtk.Plug(PlugWindowId)
NameError: global name 'gtk' is not defined

see also https://bugs.mageia.org/show_bug.cgi?id=11427 - looks like duplicates. 

And for priority and severity I agree: Release blocker and critical is more adequaate than high and major

CC: (none) => juergen.harms

*** Bug 11427 has been marked as a duplicate of this bug. ***

(In reply to Frank Griffin from comment #1)
>  Launching from a terminal as root worked, but launching
> from MCC failed reliably.
Could it be a polkit regression bug ?

CC: (none) => mageia, thierry.vignaud
Summary: System-config-printer can't start => System-config-printer can't start from drakconf

a PyGTK problem, it seems that it need to be modified to use new PyGTK

(In reply to Manuel Hiebel from comment #5)
> (In reply to Frank Griffin from comment #1)
> >  Launching from a terminal as root worked, but launching
> > from MCC failed reliably.
> Could it be a polkit regression bug ?

I really don't know.  But shouldn't all of the authorization issues be history once MCC is running ?  I mean, after the initial root password prompt, MCC is running as root, correct ?

I can't recall whether I tried MCC from a root or non-root command line to get the above data...

The problem come from our Embedded mode
system-config-printer-1.4.2-mga_custom-embedded_window.patch need to be changed
at least instead of

@@ -192,6 +196,16 @@ class GUI(GtkGUI):
 
                         domain=config.PACKAGE)
 
+        if PlugWindowId:
+            self.PrintersWindow.hide()
+            # the "vbox4" widget
+            vbox = self.PrintersWindow.get_children()[0]
+            PlugWindow = gtk.Plug(PlugWindowId)
+            gtk.Container.remove(self.PrintersWindow, vbox)
+            PlugWindow.add(vbox)
+            self.PrintersWindow.set_transient_for(PlugWindow)
+            PlugWindow.show_all()
+            self.PrintersWindow = PlugWindow

we need to write 

@@ -192,6 +196,16 @@ class GUI(GtkGUI):
 
                         domain=config.PACKAGE)
 
+        if PlugWindowId:
+            self.PrintersWindow.hide()
+            # the "vbox4" widget
+            vbox = self.PrintersWindow.get_children()[0]
+            PlugWindow = Gtk.Plug(PlugWindowId)
+            Gtk.Container.remove(self.PrintersWindow, vbox)
+            PlugWindow.add(vbox)
+            self.PrintersWindow.set_transient_for(PlugWindow)
+            PlugWindow.show_all()
+            self.PrintersWindow = PlugWindow

since now in system-config-printer.py we have
from gi.repository import Gtk

instead of previous
import gtk


But I don't know if it is enough

can someone test it ?

It's better: it doesn't crash but it still not displayed.
I've also fixed system-config-printer-1.4.2-mga_custom-system-config-printer.patch which had a similar issue.

But We still have an issue:

/usr/lib64/python2.7/site-packages/gi/overrides/Gtk.py:445: Warning: value "((GtkWindowType) 56640845)" of type 'GtkWindowType' is invalid or out of range for property 'type' of type 'GtkWindowType'
  Gtk.Window.__init__(self, type=type, **kwds)

This happens on the following line:

PlugWindow = Gtk.Plug(PlugWindowId)


Probably a regression from pygtk2 to python-gi. The issue is that we're probably the only users of Gtk.Plug using python...

Closing this one

Status: NEW => RESOLVED
Resolution: (none) => FIXED

See Bug #11546 for the python-gi issue

*** Bug 11514 has been marked as a duplicate of this bug. ***

CC: (none) => axonefr