Common Issues & Errors¶
Important
This list will be updated as issues and errors are discovered or fixed.
Note
If the information on this page fails to resolve the roadblocks that you are experiencing, please contact us using the contact information listed under the Tracktable Contacts section. Include a description of your problem, the OS and version you’re using, details of your Python/Anaconda environment, and any screenshots and error messages you can capture.
Installation¶
Pip fails to install Tracktable
Look carefully at the error message. It is most likely an error while trying to install Cartopy, a map rendering toolkit that tracktable uses to render images. This will show up as a complaint about a
GEOS
version or a request forProj 4.9.0
.The solution is to install Cartopy and its dependencies yourself before you try to install Tracktable.
If you’re using Anaconda:
The command
conda install --channel conda-forge cartopy
should resolve the issue.If you’re using
pip
:Tip
We highly recommend using Anaconda to install and manage Tracktable and its dependencies.
Install the Cartopy required dependencies
Once the dependencies are in place, the command
pip install cartopy
should succeed.
If neither of the above worked, then please refer to the Cartopy installation instructions for additional ways to install Cartopy.
After you have installed Cartopy, retry
pip install tracktable
.Python reports one of the following error messages when importing tracktable:
"Error importing C Python extension example..."
"Error importing Boost.Python example..."
"Error importing core Boost libraries..."
"Error importing Tracktable's core types library..."
This is an OS-specific compiled library issue. See the Shared Library bullet below describing this issue in more detail.
Errors about client certificates, certificate validation, or connections timing out while trying to install anything with Conda or Pip.
This is a proxy/SSL issue and outside the scope of this document.
Pip tries and fails to build Cartopy
If at all possible, please use Anaconda. The developers of Cartopy provide binary packages for Anaconda.
If you absolutely can’t use Anaconda and your operating system’s package manager does not include Cartopy, refer to the “Pip fails to install Tracktable” section above.
Build Time¶
Note
For more information about building Tracktable from source, please refer to the Installing Tracktable From Source guide. These issues and others are covered there.
Boost import targets not found
This happens when your installed version of CMake is too old for your installed version of Boost. Please upgrade CMake to at least 3.19.
Anaconda CMake package doesn’t include ccmake
This is a known bug that has been fixed in
conda-forge
but has not yet propagated to the main distribution. Installcmake
from theconda-forge
channel as follows:conda install --channel conda-forge cmake
Old Boost version found in /usr/lib or /usr/lib64
Set the
Boost_INCLUDE_DIR
variable in CMake to point to the location of the include files for your preferred Boost installation. The filenames for the compiled libraries will be updated the next time you pressc
orConfigure
.
Runtime¶
Python throwing an error similar to
“ModuleNotFoundError: No module named ‘tracktable’”
If you installed Tracktable using
pip
, this error indicates that the Python interpreter you just started is probably not the one for which you installed Tracktable. If you are using virtual environments, make sure the correct one is active.If you installed Tracktable from source, add Tracktable’s installation directory (for example,
/usr/local/Python
) to the front of yourPYTHONPATH
environment variable.
Python throwing an error similar to
“ModuleNotFoundError: No module named ‘<package name>’”
that is unrelated to TracktableIf Python can’t find a package that isn’t
tracktable
it’s possible that the package isn’t installed in the environment. If you are using virtual environments, make sure the correct one is active. If the problem persists, use one of the following commands to install the missing package:conda install --channel conda-forge <package name>
conda install <package name>
pip3 install <package name>
Tip
We highly recommend using Anaconda to install and manage tracktable and it’s dependencies.