Troubleshooting#

Logging#

Logging can be configured interactively via the Settings > Logging tab. From here, you can customise how log data is collected, stored, and used for debugging or performance analysis.

The following options are available:

  • Log Directory: The directory where log files are saved. If left empty, no log file is written.

  • Retention (days): Defines how many days logs are kept before automatic deletion.

  • Performance Logging: Enables detailed performance profiling and timing logs for key operations.

  • Log Level: Controls the level of detail outputted to the log.

The log level determines how much information is recorded. Mantid Imaging supports the following standard log levels:

Level

Description

DEBUG

Detailed developer-focused messages

INFO

General workflow steps and user actions

WARNING

Recoverable problems or fallback behaviors

ERROR

Failed operations that interrupt the workflow

CRITICAL

Severe issues such as crashes or data corruption

Note

Log file retention only applies to the currently configured log directory. If you change the directory after logs have been written, previous log files in older directories will not be automatically deleted. You may need to manage these manually.

Logging settings can also be configured manually via the QSettings configuration file.

Locations by platform:

  • Linux: ~/.config/mantidproject/Mantid Imaging.conf

  • Windows: Stored in the Windows Registry under: HKEY_CURRENT_USER\Software\mantidproject\Mantid Imaging

  • IDAaaS: Same as Linux, usually in the user’s home directory at: ~/.config/mantidproject/Mantid Imaging.conf

For example:

[logging]
log_level=DEBUG
log_dir=/tmp/mantid_imaging_logs
performance_log=true

Can’t reconstruct with FBP_CUDA and SIRT_CUDA#

These algorithms have 2 requirements:

  • Having a CUDA-compatible graphics card (GPU)

  • Having the CUDA Runtime 12.9 libraries installed

Having a CUDA-compatible graphics card#

Please check that your GPU is on the list of compatible GPUs: https://developer.nvidia.com/cuda-gpus

Having the CUDA Runtime 12.9 libraries installed#

Please install the CUDA Runtime version 12.9 binaries from https://developer.nvidia.com/cuda-downloads?target_os=Windows&target_arch=x86_64&target_version=11&target_type=exe_network

Reinstalling CUDA 12.9 on Linux can be quickly done with this script:

# uninstall nvidia driver and cuda driver
nvidia-uninstall && /usr/local/cuda/bin/cuda-uninstaller

# download the CUDA installer
wget https://developer.download.nvidia.com/compute/cuda/12.9.0/local_installers/cuda_12.9.0_550.54.14_linux.run -O ~/Downloads/cuda_12.9.0_550.54.14_linux.run

# install GPU driver and CUDA Toolkit from CLI silently
sudo bash ~/Downloads/cuda_12.9.0_550.54.14_linux.run  --silent --driver --toolkit

Specific Errors#

SystemError: <built-in function connectSlotsByName> returned a result with an error set#

This means that the PyQt package is missing or an old version (it must be newer than PyQt5==5.13.2).

Suggested Fix#

Install PyQt5 and pyqtgraph with pip install pyqt5==5.15 pyqtgraph==0.12.

qt.qpa.plugin: Could not load the Qt platform plugin “xcb” in “” even though it was found#

This might mean that some system libraries are missing.

Suggested Fix#

Try installing apt install libxcb-xinerama0.

qt.qpa.xcb: could not connect to display…#

Missing DISPLAY variable. This means the application cannot find the display to show itself on.

Suggested Fix#

You can set this via export DISPLAY=:N where :N should be the number of your display.

IDAaaS Terminal Error Messages#

Errors from the terminal when Mantid Imaging is launched, such as:

ERROR: ld.so: object 'libdlfaker.so' from LD_PRELOAD cannot be preloaded: ignored.

These are harmless and can be ignored.

UserWarning: CUDA path could not be detected. Set CUDA_PATH environment variable if CuPy failed to load#

This warning likely indicates that the CUDA_PATH environment variable is not set correctly, or that the CUDA Toolkit 12.9 is not installed on your system.

If Mantid Imaging is running correctly, you can ignore this warning. However, if you are experiencing issues with CUDA-based algorithms such as PDHG, you may need to set the CUDA_PATH environment variable to the path where CUDA Toolkit 12.9 is installed on your machine. Instructions for setting environment variables can be found below under the ImportError: DLL load failed while importing astra_c section.

ImportError: DLL load failed while importing astra_c: The specified module could not be found#

This error usually means that astra-toolbox is not installed correctly or the CUDA Runtime libraries are missing.

By default, the CUDA-version of astra-toolbox is installed during environment setup, HOWEVER it is possible for the incorrect version to be installed for non-NVIDIA GPUs or for machines without a GPU. You can check which version is installed by running the command mamba list astra-toolbox in a terminal and verifying that the version installed is the python or cuda version of the package. If the CUDA-version is installed, but you do not have a compatible GPU or the CUDA Runtime libraries installed, you will likely encounter this error.

If you are still experiencing this error after verifying that you have a compatible GPU and the CUDA Runtime libraries installed, it is possible that CUDA Toolkit 12.9 is not installed correctly or the environment variable CUDA_PATH is not set correctly.

You can check if Windows can find your CUDA device drivers are present by opening a terminal and entering the command: nvidia-smi you should see an output similar to the following:

Example output from nvidia-smi:

NVIDIA-SMI 576.57               Driver Version: 576.57         CUDA Version: 12.9
+---------------------------------------+------------------------+----------------------+
| GPU  Name                Driver-Model | Bus-Id          Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf        Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |
|                                       |                        |               MIG M. |
+=======================================+========================+======================+
|   0  NVIDIA GeForce GTX 1080 Ti WDDM  |   00000000:9E:00.0 Off |                  N/A |
| 20%   29C    P8           15W /  250W |     502MiB /  11264MiB |      3%      Default |
|                                       |                        |                  N/A |
+---------------------------------------+------------------------+----------------------+

Common causes include:

  • When the CUDA-version of astra-toolbox is installed on a machine that does not have a GPU.

  • When the Python-version of astra-toolbox is installed on a machine that has a GPU but does not have the correct CUDA Runtime libraries installed.

  • Possibly after upgrading from Windows 10 to Windows 11, where the CUDA device drivers are not compatible with the new OS or have been misplaced or removed.

Troubleshooting Steps

  1. Check for a CUDA-Compatible GPU

    • Visit the NVIDIA CUDA GPUs page to confirm your GPU is compatible.

    • If you do not have a compatible GPU, install the Python-version of astra-toolbox instead of the CUDA-version.

  2. Verify astra-toolbox Version

    • Run mamba list astra-toolbox in a terminal.

    • Ensure you have the correct version:

  • Python-version:

    • Pattern: astra-toolbox=2.1*=py*

    • Example: astra-toolbox=2.1*=py310h7b2d6b3_0

  • CUDA-version:

    • Pattern: astra-toolbox=2.1*=py*_cuda*

    • Example: astra-toolbox=2.1.*=py310h7b2d6b3_0_cuda105

Note

  • If you have a CUDA-compatible GPU but the Python-version of astra-toolbox is installed, you can install the CUDA-version by running the command: mamba install astra-toolbox=2.1.*=py310h7b2d6b3_0_cuda105 in a terminal.

  • If you are still experiencing this issue in the instance where you have a CUDA-compatible GPU and the CUDA-version of astra-toolbox is installed, it’s likely that the issue is related to your CUDA Runtime libraries.

  1. Check CUDA Toolkit Installation and Environment Variable

    • Run nvcc --version in a terminal to check if the CUDA Toolkit 12.9 is installed and accessible i.e. your CUDA_PATH environment variable has been set correctly.

    • If this fails, reinstall the CUDA Toolkit 12.9.

    • Set the environment variable CUDA_PATH to your CUDA installation path (e.g., C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.9\bin):

      1. Use Windows Search to find and open “Edit environment variables for your account” .

      2. Click “New” under user variables.

      3. Set the variable name to CUDA_PATH and the value to your CUDA installation path.

      4. Restart your machine.

      5. Open a new terminal and run echo %CUDA_PATH% to confirm.

  2. Rebuild Your Developer Environment

    • After confirming CUDA is installed and CUDA_PATH is set, run python3 ./setup.py create_dev_env to ensure the correct astra-toolbox version and CUDA libraries are available.

On this page
Show Source