Bug#895008: kicad: can't start pcbnew

[Dave Platt]

The next shoe seems to have dropped... I just did a dist-upgrade to my
Debian testing distribution on my laptop, and kicad was removed
(although kicad-common remains).  Looking at the Debian packages list
confirms that the kicad package has been removed from buster.

I'll keep my eye on the unstable distribution and hope to pull KiCad5
from there when it arrives.

Bug#895008: kicad: can't start pcbnew

[Fabián Inostroza]

I tested the package from experimental and unfortunately eeschema is
too slow when compiled against wxgtk (gtk3). It is a know problem when
wxgtk is compiled against gtk3.
So for now I'm using the mentioned workaround (solves my problem).

2018-04-14 4:59 GMT-03:00 Carsten Schoenert :
> Hello Fabián,
>
> Am 14.04.18 um 08:42 schrieb Fabián Inostroza:
>> Hi,
>>
>> I'm running debian testing and the problem is there also, I'm using
>> testing packages only.
>> The package python-wxgtk3.0 was updated to 3.0.2.0+dfsg-7 on
>> 2018-04-10, reverting to 3.0.2.0+dfsg-6 fixes the problem.
>
> no, it solves the problem not. This is just a workaround to be able to
> get the existing old packages in testing working again. The current
> state in testing for KiCad is RC buggy and once someone is open up a bug
> report about this the kicad packages will get removed from testing if we
> can't get rid of such RC things before.
>
> As written in this report I'm not intending to spend any work on the old
> version of KiCad in testing and use my time for getting the next version
> of KiCad packaged. I have no time to do both.
>
> Once the RC2 of KiCad5 is released I will upload this to unstable
> instead of experimental. So I want to suggest to test the packages in
> experimental right now as they will get into unstable in the near future.
>
> Otherwise it's currently better to use stable if you want KiCad to use
> on a stable base. But without testers and bug triage on KiCad5 it will
> take a long time to get KiCad5 into testing.
>
> --
> Regards
> Carsten

Bug#895008: kicad: can't start pcbnew

[Carsten Schoenert]

Hello Fabián,

Am 14.04.18 um 08:42 schrieb Fabián Inostroza:
> Hi,
> 
> I'm running debian testing and the problem is there also, I'm using
> testing packages only.
> The package python-wxgtk3.0 was updated to 3.0.2.0+dfsg-7 on
> 2018-04-10, reverting to 3.0.2.0+dfsg-6 fixes the problem.

no, it solves the problem not. This is just a workaround to be able to
get the existing old packages in testing working again. The current
state in testing for KiCad is RC buggy and once someone is open up a bug
report about this the kicad packages will get removed from testing if we
can't get rid of such RC things before.

As written in this report I'm not intending to spend any work on the old
version of KiCad in testing and use my time for getting the next version
of KiCad packaged. I have no time to do both.

Once the RC2 of KiCad5 is released I will upload this to unstable
instead of experimental. So I want to suggest to test the packages in
experimental right now as they will get into unstable in the near future.

Otherwise it's currently better to use stable if you want KiCad to use
on a stable base. But without testers and bug triage on KiCad5 it will
take a long time to get KiCad5 into testing.

-- 
Regards
Carsten

Bug#895008: kicad: can't start pcbnew

[Fabián Inostroza]

Hi,

I'm running debian testing and the problem is there also, I'm using
testing packages only.
The package python-wxgtk3.0 was updated to 3.0.2.0+dfsg-7 on
2018-04-10, reverting to 3.0.2.0+dfsg-6 fixes the problem.

Bug#895008: kicad: can't start pcbnew

[Leonardo Canducci]

I've tried kicad 5 RC2 and pcbnew works fine with modern toolset
(sid+gnome+xorg+intel graphics).

2018-04-08 8:00 GMT+02:00 Carsten Schoenert :

> Hello Aurelien,
>
> On Sat, Apr 07, 2018 at 11:37:25PM +0200, Aurelien Jacobs wrote:
> ...
> > I've tested this new build, and I can now start pcbnew including with
> > latest version python-wxgtk3.0, so this seems to fix the issue. Thanks !
>
> excellent, thanks for testing, I also experienced no segfaulting issues
> any more while starting.
>
> > Now this has uncovered another GTK+3 related issue for me in pcbnew.
> > If I enable "Modern Toolset (Accelerated)" in the Preferences menu,
> > pcbnew segfaults. It looks like it might be related to using
> > wxGLCanvasX11 (ie. calling libGLX and libX11) even though this GTK+3
> > build is now using Wayland instead of X11.
>
> Hmm, I've no deeper knowledge about the internals of wayland vs. X11 so
> I probably can't help much here.
> It's known that pcbnew will crash under Wayland and the Modern Toolset
> in some cases. Please note the information on
>
> http://kicad-pcb.org/help/known-system-related-issues/#_wayland
>
> As you got a informative backtrace it mayed be a good idea to add this
> to the bug report mentioned on the above linked section. As this is now
> a different problem to the original bug report here I'd like close this
> report by a upload of this rebuilded version of src:kicad and create
> later a new report with the backtrace information from you.
>
> If you could spend some time to dig into this new issue under wayland
> I'd really appreciate this!
>
> Back to the origin of this report.
> Upstream is planning to release a RC2 of KiCad 5 with a string freeze.
> The feature freeze was done with RC1 so only bugfixing will be done on
> KiCad. I plan to upload these RC2 related version to unstable to get a
> broader base for testing.
> I don't have plans to upload a rebuild for KiCad 4.0.7 in unstable.
>
> Regards
> Carsten
>



-- 
Leonardo Canducci

Bug#895008: kicad: can't start pcbnew

[Aurelien Jacobs]

On Sun, Apr 08, 2018 at 08:00:52AM +0200, Carsten Schoenert wrote:
> Hello Aurelien,
> 
> On Sat, Apr 07, 2018 at 11:37:25PM +0200, Aurelien Jacobs wrote:
> ... 
> > I've tested this new build, and I can now start pcbnew including with
> > latest version python-wxgtk3.0, so this seems to fix the issue. Thanks !
> 
> excellent, thanks for testing, I also experienced no segfaulting issues
> any more while starting.
> 
> > Now this has uncovered another GTK+3 related issue for me in pcbnew.
> > If I enable "Modern Toolset (Accelerated)" in the Preferences menu,
> > pcbnew segfaults. It looks like it might be related to using
> > wxGLCanvasX11 (ie. calling libGLX and libX11) even though this GTK+3
> > build is now using Wayland instead of X11.
> 
> Hmm, I've no deeper knowledge about the internals of wayland vs. X11 so
> I probably can't help much here.
> It's known that pcbnew will crash under Wayland and the Modern Toolset
> in some cases. Please note the information on
> 
> http://kicad-pcb.org/help/known-system-related-issues/#_wayland
> 
> As you got a informative backtrace it mayed be a good idea to add this
> to the bug report mentioned on the above linked section. As this is now
> a different problem to the original bug report here I'd like close this
> report by a upload of this rebuilded version of src:kicad and create
> later a new report with the backtrace information from you.

Sure, this is the way to go.

> If you could spend some time to dig into this new issue under wayland
> I'd really appreciate this!

This is an upstream issue. It is tracked here for kicad:
https://bugs.launchpad.net/kicad/+bug/1755360

But the issue is actually in wxWidgets and is tracked here:
https://trac.wxwidgets.org/ticket/17702

This is not easy to fix, so we will probably have to live with it for
quite some time. It might be good to document this limitation in the
README.Debian the same way kicad just documented it here:
https://github.com/KiCad/kicad-website/pull/276/commits/f1adadb744819ff8fd164ef0446a140f7fd23586
And maybe also document the solution of running:
  GDK_BACKEND=x11 kicad
for those who really want accelerated backend.

> Back to the origin of this report.
> Upstream is planning to release a RC2 of KiCad 5 with a string freeze.
> The feature freeze was done with RC1 so only bugfixing will be done on
> KiCad. I plan to upload these RC2 related version to unstable to get a
> broader base for testing. 
> I don't have plans to upload a rebuild for KiCad 4.0.7 in unstable.

Sounds OK to me, but it might be good in the meantime to upload your
build using libwxgtk3.0-gtk3-dev in experimental (it all depends if
RC2 will be released in just a few days, or several weeks).

Thanks agains.
--
Aurel

Bug#895008: kicad: can't start pcbnew

[Carsten Schoenert]

Hello Aurelien,

On Sat, Apr 07, 2018 at 11:37:25PM +0200, Aurelien Jacobs wrote:
... 
> I've tested this new build, and I can now start pcbnew including with
> latest version python-wxgtk3.0, so this seems to fix the issue. Thanks !

excellent, thanks for testing, I also experienced no segfaulting issues
any more while starting.

> Now this has uncovered another GTK+3 related issue for me in pcbnew.
> If I enable "Modern Toolset (Accelerated)" in the Preferences menu,
> pcbnew segfaults. It looks like it might be related to using
> wxGLCanvasX11 (ie. calling libGLX and libX11) even though this GTK+3
> build is now using Wayland instead of X11.

Hmm, I've no deeper knowledge about the internals of wayland vs. X11 so
I probably can't help much here.
It's known that pcbnew will crash under Wayland and the Modern Toolset
in some cases. Please note the information on

http://kicad-pcb.org/help/known-system-related-issues/#_wayland

As you got a informative backtrace it mayed be a good idea to add this
to the bug report mentioned on the above linked section. As this is now
a different problem to the original bug report here I'd like close this
report by a upload of this rebuilded version of src:kicad and create
later a new report with the backtrace information from you.

If you could spend some time to dig into this new issue under wayland
I'd really appreciate this!

Back to the origin of this report.
Upstream is planning to release a RC2 of KiCad 5 with a string freeze.
The feature freeze was done with RC1 so only bugfixing will be done on
KiCad. I plan to upload these RC2 related version to unstable to get a
broader base for testing. 
I don't have plans to upload a rebuild for KiCad 4.0.7 in unstable.

Regards
Carsten

Bug#895008: kicad: can't start pcbnew

[Aurelien Jacobs]

On Sat, Apr 07, 2018 at 09:33:08PM +0200, Carsten Schoenert wrote:
> Hi,
> 
> please test a rebuild of KiCad 5.0.05.0.0~rc1+dfsg1+20180318 which is
> using libwxgtk3.0-gtk3-dev to get a proper GTK3 bindings.

I've tested this new build, and I can now start pcbnew including with
latest version python-wxgtk3.0, so this seems to fix the issue. Thanks !

Now this has uncovered another GTK+3 related issue for me in pcbnew.
If I enable "Modern Toolset (Accelerated)" in the Preferences menu,
pcbnew segfaults. It looks like it might be related to using
wxGLCanvasX11 (ie. calling libGLX and libX11) even though this GTK+3
build is now using Wayland instead of X11.

Here is the backtrace I got:

Thread 1 "kicad" received signal SIGSEGV, Segmentation fault.
0x723aac49 in XQueryExtension () from 
/usr/lib/x86_64-linux-gnu/libX11.so.6
(gdb) disassemble $pc-32,$pc+32
Dump of assembler code from 0x723aac29 to 0x723aac69:
   0x723aac29 : sub$0x38,%rsp
   0x723aac2d : mov%fs:0x28,%rax
   0x723aac36 : mov%rax,0x28(%rsp)
   0x723aac3b : xor%eax,%eax
   0x723aac3d : mov0x968(%rdi),%rax
   0x723aac44 : test   %rax,%rax
   0x723aac47 : je 0x723aac4b 

=> 0x723aac49 : callq  *(%rax)
   0x723aac4b : mov$0x8,%edx
   0x723aac50 : mov$0x62,%esi
   0x723aac55 : mov%rbx,%rdi
   0x723aac58 : callq  0x7238db90 
<_XGetRequest@plt>
   0x723aac5d : test   %rbp,%rbp
   0x723aac60 : mov%rax,%r15
   0x723aac63 : je 0x723aad10 

End of assembler dump.
(gdb) bt
#0  0x723aac49 in XQueryExtension () at 
/usr/lib/x86_64-linux-gnu/libX11.so.6
#1  0x7fffef2ea3b2 in  () at /usr/lib/x86_64-linux-gnu/libGLX.so.0
#2  0x7fffef2e6415 in glXQueryVersion () at 
/usr/lib/x86_64-linux-gnu/libGLX.so.0
#3  0x77bcfe95 in wxGLCanvasX11::GetGLXVersion() () at 
/usr/lib/x86_64-linux-gnu/libwx_gtk3u_gl-3.0.so.0
#4  0x77bd0535 in wxGLCanvasX11::ConvertWXAttrsToGL(int const*, int*, 
unsigned long) () at /usr/lib/x86_64-linux-gnu/libwx_gtk3u_gl-3.0.so.0
#5  0x77bd0d68 in wxGLCanvasX11::InitXVisualInfo(int const*, 
__GLXFBConfigRec***, XVisualInfo**) () at 
/usr/lib/x86_64-linux-gnu/libwx_gtk3u_gl-3.0.so.0
#6  0x77bd1e70 in wxGLCanvas::Create(wxWindow*, int, wxPoint const&, 
wxSize const&, long, wxString const&, int const*, wxPalette const&) () at 
/usr/lib/x86_64-linux-gnu/libwx_gtk3u_gl-3.0.so.0
#7  0x77bd2013 in wxGLCanvas::wxGLCanvas(wxWindow*, int, int const*, 
wxPoint const&, wxSize const&, long, wxString const&, wxPalette const&) () at 
/usr/lib/x86_64-linux-gnu/libwx_gtk3u_gl-3.0.so.0
#8  0x7fffd1aedbd9 in HIDPI_GL_CANVAS::HIDPI_GL_CANVAS(wxWindow*, int, int 
const*, wxPoint const&, wxSize const&, long, wxString const&, wxPalette const&) 
(this=0x585c4ea0, parent=, id=, 
attribList=, pos=..., size=..., style=8192, name=..., 
palette=...) at ./common/gal/hidpi_gl_canvas.cpp:38
#9  0x7fffd1b09285 in 
KIGFX::OPENGL_GAL::OPENGL_GAL(KIGFX::GAL_DISPLAY_OPTIONS&, wxWindow*, 
wxEvtHandler*, wxEvtHandler*, wxString const&) (this=0x585c4bb0, 
aDisplayOptions=..., aParent=0x57d7f120, aMouseListener=0x57d7f120, 
aPaintListener=0x57d7f120, aName=...) at 
./common/gal/opengl/opengl_gal.cpp:74
#10 0x7fffd1ae4b84 in 
EDA_DRAW_PANEL_GAL::SwitchBackend(EDA_DRAW_PANEL_GAL::GAL_TYPE) 
(this=this@entry=0x57d7f120, aGalType=EDA_DRAW_PANEL_GAL::GAL_TYPE_OPENGL) 
at ./common/draw_panel_gal.cpp:350
#11 0x7fffd189a01e in 
PCB_DRAW_PANEL_GAL::SwitchBackend(EDA_DRAW_PANEL_GAL::GAL_TYPE) 
(this=0x57d7f120, aGalType=) at 
./pcbnew/pcb_draw_panel_gal.cpp:396
#12 0x7fffd19e997f in 
EDA_DRAW_FRAME::SwitchCanvas(EDA_DRAW_PANEL_GAL::GAL_TYPE) 
(this=0x57d652b0, aCanvasType=EDA_DRAW_PANEL_GAL::GAL_TYPE_OPENGL) at 
./common/draw_frame.cpp:1195
#13 0x7fffd147d249 in PCB_EDIT_FRAME::OnSwitchCanvas(wxCommandEvent&) 
(this=0x57d652b0, aEvent=...) at ./pcbnew/pcb_edit_frame.cpp:1215
#14 0x7654a8ce in 
wxEvtHandler::ProcessEventIfMatchesId(wxEventTableEntryBase const&, 
wxEvtHandler*, wxEvent&) () at /usr/lib/x86_64-linux-gnu/libwx_baseu-3.0.so.0
#15 0x7654a9d3 in wxEventHashTable::HandleEvent(wxEvent&, 
wxEvtHandler*) () at /usr/lib/x86_64-linux-gnu/libwx_baseu-3.0.so.0
#16 0x7654ad9b in wxEvtHandler::TryHereOnly(wxEvent&) () at 
/usr/lib/x86_64-linux-gnu/libwx_baseu-3.0.so.0
#17 0x7fffd1a0e1eb in EDA_BASE_FRAME::ProcessEvent(wxEvent&) 
(this=0x57d652b0, aEvent=...) at 

Bug#895008: kicad: can't start pcbnew

[Carsten Schoenert]

Hi,

please test a rebuild of KiCad 5.0.05.0.0~rc1+dfsg1+20180318 which is
using libwxgtk3.0-gtk3-dev to get a proper GTK3 bindings.

https://people.debian.org/~tijuca/kicad/

Regards
Carsten

Bug#895008: kicad: can't start pcbnew

[Carsten Schoenert]

Hello Leonardo and Aurel,

On Fri, Apr 06, 2018 at 03:39:30PM +0200, Leonardo Canducci wrote:
> Just to clarify: I'm running sid and gnome with debian repository only.
> Kicad did work a couple of months ago.
> 
> Downgrading python-wxgtk3.0 fixes the problem here too.
> 
> Leonardo
> 
> 2018-04-06 15:19 GMT+02:00 Aurelien Jacobs :
> 
> > Hello,
> >
> > I can confirm this issue.
> >
> > Since I upgraded my sid install today, pcbnew fails to start with those
> > kind of messages :
...
> > I reproduced this issue on 2 different machines (both updated to latest
> > sid today).
> >
> > It seems to be realted to latest version of python-wxgtk3.0.
> > When I downgrade this package to 3.0.2.0+dfsg-6 (from buster) pcbnew
> > starts working again.

thanks for checking this, I'm on testing and there isn't this issue
happen.
src:wxpython3.0 is depending on src:wxwidgets3.0-gtk3 and this package
is on a way of a transition [1]. So one day kicad would be binNMU'ed. I'll
have a look at this on the weekend.

[1] https://release.debian.org/transitions/html/wxwidgets3.0-gtk3.html

Regards
Carsten

Bug#895008: kicad: can't start pcbnew

[Leonardo Canducci]

Just to clarify: I'm running sid and gnome with debian repository only.
Kicad did work a couple of months ago.

Downgrading python-wxgtk3.0 fixes the problem here too.

Leonardo

2018-04-06 15:19 GMT+02:00 Aurelien Jacobs :

> Hello,
>
> I can confirm this issue.
>
> Since I upgraded my sid install today, pcbnew fails to start with those
> kind of messages :
>
> ../src/common/object.cpp(251): assert "classTable->Get(m_className) ==
> NULL" failed in Register(): Class "wxCommandEvent" already in RTTI table -
> have you used IMPLEMENT_DYNAMIC_CLASS() multiple times or linked some
> object file twice)?
> ../src/common/object.cpp(251): assert "classTable->Get(m_className) ==
> NULL" failed in Register(): Class "wxNotifyEvent" already in RTTI table -
> have you used IMPLEMENT_DYNAMIC_CLASS() multiple times or linked some
> object file twice)?
> ../src/common/object.cpp(251): assert "classTable->Get(m_className) ==
> NULL" failed in Register(): Class "wxScrollEvent" already in RTTI table -
> have you used IMPLEMENT_DYNAMIC_CLASS() multiple times or linked some
> object file twice)?
> ../src/common/object.cpp(251): assert "classTable->Get(m_className) ==
> NULL" failed in Register(): Class "wxScrollWinEvent" already in RTTI table
> - have you used IMPLEMENT_DYNAMIC_CLASS() multiple times or linked some
> object file twice)?
> .
>
> I reproduced this issue on 2 different machines (both updated to latest
> sid today).
>
> It seems to be realted to latest version of python-wxgtk3.0.
> When I downgrade this package to 3.0.2.0+dfsg-6 (from buster) pcbnew
> starts working again.
>
> --
> Aurel
>



-- 
Leonardo Canducci

Bug#895008: kicad: can't start pcbnew

[Aurelien Jacobs]

Hello,

I can confirm this issue.

Since I upgraded my sid install today, pcbnew fails to start with those
kind of messages :

../src/common/object.cpp(251): assert "classTable->Get(m_className) == NULL" 
failed in Register(): Class "wxCommandEvent" already in RTTI table - have you 
used IMPLEMENT_DYNAMIC_CLASS() multiple times or linked some object file twice)?
../src/common/object.cpp(251): assert "classTable->Get(m_className) == NULL" 
failed in Register(): Class "wxNotifyEvent" already in RTTI table - have you 
used IMPLEMENT_DYNAMIC_CLASS() multiple times or linked some object file twice)?
../src/common/object.cpp(251): assert "classTable->Get(m_className) == NULL" 
failed in Register(): Class "wxScrollEvent" already in RTTI table - have you 
used IMPLEMENT_DYNAMIC_CLASS() multiple times or linked some object file twice)?
../src/common/object.cpp(251): assert "classTable->Get(m_className) == NULL" 
failed in Register(): Class "wxScrollWinEvent" already in RTTI table - have you 
used IMPLEMENT_DYNAMIC_CLASS() multiple times or linked some object file twice)?
.

I reproduced this issue on 2 different machines (both updated to latest
sid today).

It seems to be realted to latest version of python-wxgtk3.0.
When I downgrade this package to 3.0.2.0+dfsg-6 (from buster) pcbnew
starts working again.

--
Aurel

Bug#895008: kicad: can't start pcbnew

[Carsten Schoenert]

Hello Leonardo,

On Fri, Apr 06, 2018 at 11:31:38AM +0200, Leonardo Canducci wrote:
> 2018-04-06 11:03 GMT+02:00 Carsten Schoenert :
> > As you also have problems with the packages from experimental too I
> > assume you have more a problem with local modification than with the
> > KiCad packages itself.
> >
> 
> Is there something I can do to track down what's wrong?

sure, but this depends on your local system and setup.
I don't know if you probaly have something modified on your installed
packages or installed from a different source or installed from source
anyhow as you dot gave some information about such adjustments.
There are various of possibilities that can provoke such a behavour.

The 4.0.7 is out there for a long time and you are the first person with
a bug report for long time, so it's more likely in my eyes that
something must be skrewed up on your system. You will need to use tools
like strace and gdb to track down which library is bringing up that
mess.

Can this issue be reproduced on other machines? At last I can't on my
two PCs here.
 
> > You also have apparmor enabled, did the problems still exists if you
> > have apparmor disabled? Are any apparmor related logging entries
> > vissible that let you assume apparmor could be the root of the problems?
> >
> 
> Not that I can see. I didn't disable apparmor but I can't find anything
> related in
> aa-status output (attached).

This doesn't help that much, aa-status tells you just which profiles are
enabled.

You need to look at the syslog to see if the kernel or the apprmor
process is complaining about something.

https://wiki.debian.org/AppArmor/Debug

Regards
Carsten

Bug#895008: kicad: can't start pcbnew

[Leonardo Canducci]

2018-04-06 11:03 GMT+02:00 Carsten Schoenert :

> Hello Leonardo,
>
> On Fri, Apr 06, 2018 at 09:47:59AM +0200, Leonardo Canducci wrote:
> > Assertion failed!
> > ../src/common/object.cpp(251): assert "classTable->Get(m_className) ==
> NULL" failed in Register(): Class "wxCommandEvent" already in RTTI table -
> have you used IMPLEMENT_DYNAMIC_CLASS() multiple times or linked some
> object file twice)?
>
> This looks a bit like problems with python-wxgtk3.0. A error message
> like this mostlx comes from using the wrong symbol usage due different
> linking of wxpython vs wxwidgets (GTK2 vs. GTK3).
>
> To dig more into this some inspection with GDB would be needed around
> the symbols around the assertation.
>

> > Many more errors come after, all reporting
> > "../src/common/object.cpp(251): assert "classTable->Get(m_className) ==
> NULL"
> >
> > Anyway pcbnew doesn't start.
> >
> > I've tried installing kicad 5 from experimental but pcbnew won't start
> > either.
>
> As you also have problems with the packages from experimental too I
> assume you have more a problem with local modification than with the
> KiCad packages itself.
>

Is there something I can do to track down what's wrong?


> You also have apparmor enabled, did the problems still exists if you
> have apparmor disabled? Are any apparmor related logging entries
> vissible that let you assume apparmor could be the root of the problems?
>

Not that I can see. I didn't disable apparmor but I can't find anything
related in
aa-status output (attached).

-- 
Leonardo Canducci
apparmor module is loaded.
24 profiles are loaded.
22 profiles are in enforce mode.
   /usr/bin/evince
   /usr/bin/evince-previewer
   /usr/bin/evince-previewer//sanitized_helper
   /usr/bin/evince-thumbnailer
   /usr/bin/evince-thumbnailer//sanitized_helper
   /usr/bin/evince//sanitized_helper
   /usr/bin/man
   /usr/lib/cups/backend/cups-pdf
   /usr/lib/telepathy/mission-control-5
   /usr/lib/telepathy/telepathy-*
   /usr/lib/telepathy/telepathy-*//pxgsettings
   /usr/lib/telepathy/telepathy-*//sanitized_helper
   /usr/lib/telepathy/telepathy-ofono
   /usr/sbin/cups-browsed
   /usr/sbin/cupsd
   /usr/sbin/cupsd//third_party
   /usr/sbin/ntpd
   libreoffice-senddoc
   libreoffice-soffice//gpg
   libreoffice-xpdfimport
   man_filter
   man_groff
2 profiles are in complain mode.
   libreoffice-oopslash
   libreoffice-soffice
5 processes have profiles defined.
5 processes are in enforce mode.
   /usr/lib/telepathy/mission-control-5 (1511) 
   /usr/lib/telepathy/telepathy-* (1969) 
   /usr/sbin/cups-browsed (16800) 
   /usr/sbin/cupsd (16641) 
   /usr/sbin/ntpd (717) 
0 processes are in complain mode.
0 processes are unconfined but have a profile defined.

Bug#895008: kicad: can't start pcbnew

[Carsten Schoenert]

Hello Leonardo,

On Fri, Apr 06, 2018 at 09:47:59AM +0200, Leonardo Canducci wrote:
> Package: kicad
> Version: 4.0.7+dfsg1-1
> Severity: important
> 
> Dear Maintainer,
> when running pcbnew within kicad or on the command line I get lots of
> errors lihe this one (both in a dialog and in the shell):

currently I can't reproduce this on my side. No matter if run KiCad from
the UI or by the command line. I don't see any issues anythere.

> Assertion failed!
> ../src/common/object.cpp(251): assert "classTable->Get(m_className) == NULL" 
> failed in Register(): Class "wxCommandEvent" already in RTTI table - have you 
> used IMPLEMENT_DYNAMIC_CLASS() multiple times or linked some object file 
> twice)?

This looks a bit like problems with python-wxgtk3.0. A error message
like this mostlx comes from using the wrong symbol usage due different
linking of wxpython vs wxwidgets (GTK2 vs. GTK3).

To dig more into this some inspection with GDB would be needed around
the symbols around the assertation.

> Many more errors come after, all reporting 
> "../src/common/object.cpp(251): assert "classTable->Get(m_className) == NULL"
> 
> Anyway pcbnew doesn't start.
> 
> I've tried installing kicad 5 from experimental but pcbnew won't start
> either.  

As you also have problems with the packages from experimental too I
assume you have more a problem with local modification than with the
KiCad packages itself.
You also have apparmor enabled, did the problems still exists if you
have apparmor disabled? Are any apparmor related logging entries
vissible that let you assume apparmor could be the root of the problems?

Regards
Carsten

Bug#895008: kicad: can't start pcbnew

[Leonardo Canducci]

Package: kicad
Version: 4.0.7+dfsg1-1
Severity: important

Dear Maintainer,
when running pcbnew within kicad or on the command line I get lots of
errors lihe this one (both in a dialog and in the shell):

Assertion failed!
../src/common/object.cpp(251): assert "classTable->Get(m_className) == NULL" 
failed in Register(): Class "wxCommandEvent" already in RTTI table - have you 
used IMPLEMENT_DYNAMIC_CLASS() multiple times or linked some object file twice)?

Many more errors come after, all reporting 
"../src/common/object.cpp(251): assert "classTable->Get(m_className) == NULL"

Anyway pcbnew doesn't start.

I've tried installing kicad 5 from experimental but pcbnew won't start
either.  

A backtrace is attached for the first error.

Leonardo


-- System Information:
Debian Release: buster/sid
  APT prefers unstable
  APT policy: (500, 'unstable'), (500, 'testing'), (500, 'stable'), (1, 
'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.15.0-2-amd64 (SMP w/4 CPU cores)
Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968), LANGUAGE=C 
(charmap=ANSI_X3.4-1968)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

Versions of packages kicad depends on:
ii  kicad-common4.0.7+dfsg1-1
ii  libboost-atomic1.62.0   1.62.0+dfsg-5
ii  libboost-chrono1.62.0   1.62.0+dfsg-5
ii  libboost-context1.62.0  1.62.0+dfsg-5
ii  libboost-date-time1.62.01.62.0+dfsg-5
ii  libboost-filesystem1.62.0   1.62.0+dfsg-5
ii  libboost-iostreams1.62.01.62.0+dfsg-5
ii  libboost-locale1.62.0   1.62.0+dfsg-5
ii  libboost-program-options1.62.0  1.62.0+dfsg-5
ii  libboost-regex1.62.01.62.0+dfsg-5
ii  libboost-system1.62.0   1.62.0+dfsg-5
ii  libboost-thread1.62.0   1.62.0+dfsg-5
ii  libc6   2.27-3
ii  libcairo2   1.15.10-2
ii  libcurl37.58.0-2
ii  libgcc1 1:8-20180402-1
ii  libgl1  1.0.0+git20180308-1
ii  libglew2.0  2.0.0-5
ii  libglu1-mesa [libglu1]  9.0.0-2.1
ii  libgomp18-20180402-1
ii  libpython2.72.7.14-8
ii  libssl1.1   1.1.0h-2
ii  libstdc++6  8-20180402-1
ii  libwxbase3.0-0v53.0.4+dfsg-3
ii  libwxgtk3.0-0v5 3.0.4+dfsg-3
ii  python  2.7.14-4
ii  python-wxgtk3.0 3.0.2.0+dfsg-7

Versions of packages kicad recommends:
ii  xsltproc  1.1.29-5

Versions of packages kicad suggests:
ii  extra-xdg-menus  1.0-4
ii  kicad-doc-en 4.0.7+dfsg1-1
ii  kicad-doc-it 4.0.7+dfsg1-1

-- no debconf information
ASSERT INFO:
../src/common/object.cpp(251): assert "classTable->Get(m_className) == NULL" 
failed in Register(): Class "wxCommandEvent" already in RTTI table - have you 
used IMPLEMENT_DYNAMIC_CLASS() multiple times or linked some object file twice)?

BACKTRACE:
[1] wxClassInfo::Register()
[2] _dl_catch_exception
[3] _dl_catch_exception
[4] _dl_catch_error
[5] dlopen
[6] _PyImport_GetDynLoadFunc
[7] _PyImport_LoadDynamicModule
[8] PyImport_ImportModuleLevel
[9] PyObject_Call
[10] PyEval_CallObjectWithKeywords
[11] PyEval_EvalFrameEx
[12] PyEval_EvalCodeEx
[13] PyEval_EvalCode
[14] PyImport_ExecCodeModuleEx
[15] PyImport_ImportModuleLevel
[16] PyObject_Call
[17] PyEval_CallObjectWithKeywords
[18] PyEval_EvalFrameEx
[19] PyEval_EvalCodeEx
[20] PyEval_EvalCode
[21] PyImport_ExecCodeModuleEx
[22] PyImport_ImportModuleLevel
[23] PyObject_Call
[24] PyObject_CallFunction
[25] PyImport_Import
[26] PyImport_ImportModule
[27] PyCObject_Import
[28] wxEntry(int&, wchar_t**)
[29] __libc_start_main
[30] _start