[SOLVED]Cura binary Failed to load module canberra-gtk-modul

[NTwoO]

Since a recent update, my system fails to start up the appimage downloaded from the Cura project. Could it be related that there is not a gtk2 flag in libcanberra anymore?

The relevant binary is an appImage that can be downloaded at: https://github.com/Ultimaker/Cura/releases and the one tested with is the latest stable release https://github.com/Ultimaker/Cura/relea ... 4.AppImage

[rab0171610]

Ideally, appimages should be self contained and not dependent on any given version of an app or library to be installed.
That is not always the case.
I ran the appimage up until the login screen and did not have any issues.
Is media-libs/libcanberra-gtk3 installed on your system?
When looking at the source code the appimage uses Gtk-3 not Gtk-2 so I don't think that is related.
You should file a bug at their issues page. I did notice that all the user's debug output at:
https://github.com/Ultimaker/Cura/issues/19837
shows the error (Failed to load module "canberra-gtk-module") but that doesn't appear to be fatal nor discussed in the thread, so I assume it is not relevant. I suspect it is a typical warning that usually occurs and has no bearing on the appimage running correctly.
In other words, I have doubts that is the actual fatal error preventing the appimage from working correctly on your system.

[NTwoO]

Thanks for looking into it.

The output when launching the application shows the following, no more.

Code: Select all

~/Downloads/UltiMaker-Cura-5.8.0-linux-X64.AppImage 
Gtk-Message: 12:13:28.754: Failed to load module "canberra-gtk-module"
Fontconfig warning: "/etc/fonts/conf.avail/05-reset-dirs-sample.conf", line 6: unknown element "reset-dirs"
qt.glx: qglx_findConfig: Failed to finding matching FBConfig for QSurfaceFormat(version 4.1, options QFlags<QSurfaceFormat::FormatOption>(), depthBufferSize -1, redBufferSize 1, greenBufferSize 1, blueBufferSize 1, alphaBufferSize -1, stencilBufferSize -1, samples -1, swapBehavior QSurfaceFormat::SingleBuffer, swapInterval 1, colorSpace QColorSpace(), profile  QSurfaceFormat::CoreProfile)
qt.glx: qglx_findConfig: Failed to finding matching FBConfig for QSurfaceFormat(version 4.1, options QFlags<QSurfaceFormat::FormatOption>(), depthBufferSize -1, redBufferSize 1, greenBufferSize 1, blueBufferSize 1, alphaBufferSize -1, stencilBufferSize -1, samples -1, swapBehavior QSurfaceFormat::SingleBuffer, swapInterval 1, colorSpace QColorSpace(), profile  QSurfaceFormat::CoreProfile)
Could not initialize GLX

canberra libs

Code: Select all

eix canberra
[I] media-libs/libcanberra
     Available versions:  0.30-r8 {alsa gstreamer oss pulseaudio tdb udev ABI_MIPS="n32 n64 o32" ABI_S390="32 64" ABI_X86="32 64 x32"}
     Installed versions:  0.30-r8(01:17:01 PM 01/24/2025)(alsa pulseaudio udev -gstreamer -oss -tdb ABI_MIPS="-n32 -n64 -o32" ABI_S390="-32 -64" ABI_X86="64 -32 -x32")
     Homepage:            https://0pointer.de/lennart/projects/libcanberra/
     Description:         Portable sound event library

[I] media-libs/libcanberra-gtk3
     Available versions:  0.30 {gnome ABI_MIPS="n32 n64 o32" ABI_S390="32 64" ABI_X86="32 64 x32"}
     Installed versions:  0.30(03:17:05 PM 01/28/2025)(-gnome ABI_MIPS="-n32 -n64 -o32" ABI_S390="-32 -64" ABI_X86="64 -32 -x32")
     Homepage:            https://0pointer.de/lennart/projects/libcanberra/
     Description:         GTK3 bindings for libcanberra, the portable sound event library

Found 2 matches

Code: Select all

GTK_MODULES=canberra-gtk-module

EDIT: The link you shared does match my problem for the rest. I'll look into what people did there to get things running
EDIT2: Downgraded Mesa to the previous stable version. Now things work.

[rab0171610]

I have an Nvidia GPU. I am assuming since you were affected by the error in that link (Could not initialize GLX) that you have either an AMD GPU or Intel. I don't know enough about it so I can't say if it is feasible or even recommended, but rather than downgrade, building Mesa without the opencl use flag might be an alternative solution until they update their appimage. You can try it or someone else here that is more familiar with the Mesa ebuild may be able to explain why opencl is recommended or necessary with your particular graphics card.
I assumed this, possibly incorrectly, based on the comment on the issues link:

Code: Select all

If MESA is compiled with Rusticl the AppImage fails to start on some Intel/AMD iGPUs, so many distributions are affected!

You can downgrade (no nice solution) mesa to an older version and the AppImage will start again.

To fix this issue the code that initializes GLX should be updated to work with latest release of mesa

[NTwoO]

So the problem returned... I did some digging, but the problem still persists:

References I found are that shows similar behaviour:
Fellow penguin had the problem solved upstream
Zoom had a similar problem. I'll have to dive into this thread to see how people solved this.
An old thread with similar issues. Maybe this could be helpful, but most probably not
This small test program ulilising the Cura test works

Code: Select all

import sys
from PyQt6.QtGui import QOpenGLContext, QGuiApplication, QSurfaceFormat, QWindow, QSurface
from PyQt6.QtOpenGL import QOpenGLVersionProfile


app = QGuiApplication(sys.argv)
gl_window = QWindow()
gl_window.setSurfaceType(QSurface.SurfaceType.OpenGLSurface)
gl_window.showMinimized()
new_format = QSurfaceFormat()
print("1")
new_format.setMajorVersion(4)
print("2")
new_format.setMinorVersion(1) #also works for 2-6
print("3")
profile_ = QSurfaceFormat.OpenGLContextProfile.CoreProfile
print(profile_)
print("4")

new_format.setProfile(profile_)
print(new_format)
print("5")
new_context = QOpenGLContext()
print(new_context)
print("6")
new_context.setFormat(new_format)
print(new_context)
print("7")
if new_context is not None:
    print("8a")
    success = new_context.create()
    print("8b")
print("8")

[NTwoO]

Man! I thought I'd downgrade Mesa to one of the older versions. The one I blocked was 24.3.4-r1. That was the version that broke last time round. I thought to go down to 24.2.8, but pulling in the old ebuild showed that this is not trivial. Looking at package.use.mask I see a message that osmesa is removed since the latest mesa 25 version. Ubuntu 25.04 still rolls with Mesa 24.3.4 and this will then be the major version used for deploying Linux (and where Linux versions of Cura is running). If this is the cause, then getting Cura fixed is not going to be simple. Uninstalling Gentoo after 18 years and installing Ubuntu does feel like defeat...

The option shared in some of the threads to export QT_XCB_GL_INTEGRATION=none did not work as this just nuked the GL detection and produced a popup that acceleration was not possible. Cura uses Python 3.12 which is the same version that my system uses.

Next step, turn on debugging in Mesa. Last time downgrading was a solution, maybe there are reports in that corner why things fail this time.

[NTwoO]

So digging through the call stack produced by the error:

Code: Select all

./UltiMaker-Cura-5.10.1-linux-X64.AppImage --debug
UM/Settings/SettingFunction.py:244: DeprecationWarning: ast.Str is deprecated and will be removed in Python 3.14; use ast.Constant instead
  def visit_Str(self, node: ast.Str) -> None:
trimesh/interfaces/scad.py:28: DeprecationWarning: Use shutil.which instead of find_executable
  _scad_executable = find_executable('openscad', path=_search_path)
trimesh/interfaces/blender.py:36: DeprecationWarning: Use shutil.which instead of find_executable
  _blender_executable = find_executable('blender', path=_search_path)
trimesh/interfaces/vhacd.py:21: DeprecationWarning: Use shutil.which instead of find_executable
  _vhacd_executable = find_executable(_name, path=_search_path)
trimesh/curvature.py:12: DeprecationWarning: Please use `coo_matrix` from the `scipy.sparse` namespace, the `scipy.sparse.coo` namespace is deprecated.
  from scipy.sparse.coo import coo_matrix
trimesh/exchange/ply.py:953: DeprecationWarning: Use shutil.which instead of find_executable
  draco_encoder = find_executable('draco_encoder')
trimesh/util.py:89: DeprecationWarning: 'pkgutil.find_loader' is deprecated and slated for removal in Python 3.14; use importlib.util.find_spec() instead
  return pkgutil.find_loader(name) is not None
trimesh/exchange/binvox.py:18: DeprecationWarning: Use shutil.which instead of find_executable
  binvox_encoder = find_executable('binvox')
trimesh/path/exchange/dxf.py:1101: DeprecationWarning: Use shutil.which instead of find_executable
  _teigha = find_executable(_name)
Gtk-Message: 08:15:11.047: Failed to load module "canberra-gtk-module"

(UltiMaker-Cura:11485): dbind-WARNING **: 08:15:11.192: Couldn't connect to accessibility bus: Failed to connect to socket /root/.cache/at-spi/bus_0: Permission denied
Fontconfig warning: "/etc/fonts/conf.avail/05-reset-dirs-sample.conf", line 6: unknown element "reset-dirs"
[MainThread] UM.Application.initialize [163]: Initializing UltiMaker Cura
[MainThread] UM.Application.initialize [164]: App Version 5.10.1
[MainThread] UM.Application.initialize [165]: Api Version 8.10.0
[MainThread] UM.Application.initialize [166]: Build type None
[MainThread] UM.PluginRegistry.addPluginLocation [205]: Plugin location /tmp/.mount_UltiMa2Hwv8f/../Resources/share/uranium/plugins must be a folder.
[MainThread] UM.PluginRegistry.addPluginLocation [205]: Plugin location /tmp/.mount_UltiMa2Hwv8f/../Resources/share/cura/plugins must be a folder.
[MainThread] UM.PluginRegistry.addPluginLocation [205]: Plugin location /tmp/lib/uranium must be a folder.
[MainThread] UM.PluginRegistry.addPluginLocation [205]: Plugin location /tmp/lib64/uranium must be a folder.
[MainThread] UM.PluginRegistry.addPluginLocation [205]: Plugin location /tmp/lib32/uranium must be a folder.
[MainThread] UM.PluginRegistry.addPluginLocation [205]: Plugin location /tmp/.mount_UltiMa2Hwv8f/plugins must be a folder.
[MainThread] UM.PluginRegistry.addPluginLocation [205]: Plugin location /tmp/Resources/uranium/plugins must be a folder.
[MainThread] UM.PluginRegistry.addPluginLocation [205]: Plugin location /tmp/Resources/cura/plugins must be a folder.
[MainThread] UM.Resources.__initializeStoragePaths [481]: Initializing storage paths
[MainThread] UM.Resources.__initializeStoragePaths [491]: Config storage path is /home/frans/.config/cura/5.10
[MainThread] UM.Resources.__initializeStoragePaths [499]: Data storage path is /home/frans/.local/share/cura/5.10
[MainThread] UM.Resources.__initializeStoragePaths [511]: Cache storage path is /home/frans/.cache/cura/5.10
UM/Qt/QtApplication.py:701: DeprecationWarning: QEvent constructor is deprecated
  super().__init__(self.QtFunctionEvent)
[MainThread] UM.PackageManager.__init__ [56]: Found bundled packages JSON file: /tmp/.mount_UltiMa2Hwv8f/share/uranium/resources/bundled_packages/uranium.json
[MainThread] UM.PackageManager.__init__ [56]: Found bundled packages JSON file: /tmp/.mount_UltiMa2Hwv8f/share/cura/resources/bundled_packages/cura.json
[MainThread] UM.View.GL.OpenGLContext.detectBestOpenGLVersion [124]: Trying OpenGL context 4.1...
qt.glx: qglx_findConfig: Failed to finding matching FBConfig for QSurfaceFormat(version 4.1, options QFlags<QSurfaceFormat::FormatOption>(), depthBufferSize -1, redBufferSize 1, greenBufferSize 1, blueBufferSize 1, alphaBufferSize -1, stencilBufferSize -1, samples -1, swapBehavior QSurfaceFormat::SingleBuffer, swapInterval 1, colorSpace QColorSpace(), profile  QSurfaceFormat::CoreProfile)
qt.glx: qglx_findConfig: Failed to finding matching FBConfig for QSurfaceFormat(version 4.1, options QFlags<QSurfaceFormat::FormatOption>(), depthBufferSize -1, redBufferSize 1, greenBufferSize 1, blueBufferSize 1, alphaBufferSize -1, stencilBufferSize -1, samples -1, swapBehavior QSurfaceFormat::SingleBuffer, swapInterval 1, colorSpace QColorSpace(), profile  QSurfaceFormat::CoreProfile)
Could not initialize GLX
Fatal Python error: Aborted

Current thread 0x00007fb5bd654740 (most recent call first):
  File "UM/View/GL/OpenGLContext.py", line 36 in setContext
  File "UM/View/GL/OpenGLContext.py", line 125 in detectBestOpenGLVersion
  File "UM/Qt/QtApplication.py", line 178 in initialize
  File "cura/CuraApplication.py", line 349 in initialize
  File "UM/Application.py", line 288 in run
  File "UM/Qt/QtApplication.py", line 385 in run
  File "cura/CuraApplication.py", line 881 in run
  File "cura_app.py", line 243 in <module>

Extension modules: PyQt6.QtCore, PyQt6.QtNetwork, PyQt6.QtGui, PyQt6.QtWidgets, numpy.core._multiarray_umath, numpy.core._multiarray_tests, numpy.linalg._umath_linalg, numpy.fft._pocketfft_internal, numpy.random._common, numpy.random.bit_generator, numpy.random._bounded_integers, numpy.random._mt19937, numpy.random.mtrand, numpy.random._philox, numpy.random._pcg64, numpy.random._sfc64, numpy.random._generator, scipy._lib._ccallback_c, scipy.sparse._sparsetools, _csparsetools, scipy.sparse._csparsetools, scipy.sparse.linalg._isolve._iterative, scipy.linalg._fblas, scipy.linalg._flapack, scipy.linalg.cython_lapack, scipy.linalg._cythonized_array_utils, scipy.linalg._solve_toeplitz, scipy.linalg._decomp_lu_cython, scipy.linalg._matfuncs_sqrtm_triu, scipy.linalg.cython_blas, scipy.linalg._matfuncs_expm, scipy.linalg._decomp_update, scipy.linalg._flinalg, scipy.sparse.linalg._dsolve._superlu, scipy.sparse.linalg._eigen.arpack._arpack, scipy.sparse.csgraph._tools, scipy.sparse.csgraph._shortest_path, scipy.sparse.csgraph._traversal, scipy.sparse.csgraph._min_spanning_tree, scipy.sparse.csgraph._flow, scipy.sparse.csgraph._matching, scipy.sparse.csgraph._reordering, scipy.spatial._ckdtree, scipy._lib.messagestream, scipy.spatial._qhull, scipy.spatial._voronoi, scipy.spatial._distance_wrap, scipy.spatial._hausdorff, scipy.special._ufuncs_cxx, scipy.special._ufuncs, scipy.special._specfun, scipy.special._comb, scipy.special._ellip_harm_2, scipy.spatial.transform._rotation, _cffi_backend, PyQt6.QtOpenGL, PyQt6.QtQml, PyQt6.QtQuick, pyArcus, PyQt6.QtSvg, pyclipper._pyclipper, pynest2d, scipy.optimize._minpack2, scipy.optimize._group_columns, scipy.optimize._trlib._trlib, scipy.optimize._lbfgsb, _moduleTNC, scipy.optimize._moduleTNC, scipy.optimize._cobyla, scipy.optimize._slsqp, scipy.optimize._minpack, scipy.optimize._lsq.givens_elimination, scipy.optimize._zeros, scipy.optimize.__nnls, scipy.optimize._highs.cython.src._highs_wrapper, scipy.optimize._highs._highs_wrapper, scipy.optimize._highs.cython.src._highs_constants, scipy.optimize._highs._highs_constants, scipy.linalg._interpolative, scipy.optimize._bglu_dense, scipy.optimize._lsap, scipy.optimize._direct, shapely.lib, shapely._geos, shapely._geometry_helpers, scipy.ndimage._nd_image, _ni_label, scipy.ndimage._ni_label (total: 88)
Aborted

I think the problem is that no configs are given by glxChooseFBConfig in the call qglx_findConfig. The sources are here.

The rest of that flow looks fine and looking at hte QSurfaceFormat being matched, things just don't such a high spec. My native test returns matches up to 4.6 and 4.1/4.0 also returns a match.

[NTwoO]

There is another lead... I thought I'd start a VM for running Cura. There are some prints that just have to go out the door. Starting VirtualBox produced a list of errors, also mentioning the dbus. There was also a dbus error in the Cura logs. Time to start scouring the forum for other people with dbus errors.

Code: Select all

Qt WARNING: Session DBus not running.
Qt WARNING: Application will not react to setting changes.
 Check your DBus installation.

[NTwoO]

OK. Now I in any case have the software running in a VM with Ubuntu 24.04..

[NTwoO]

OK! I squashed it!! No moving to Ubuntu for the time being!!!!

Removing llvm from the use flags of mesa solved the problem. I don't know why, but it does. I also added z3 to llvm, but not sure if this was the cause. I think that mesa's involvement with resolving the framebuffer was more important.