• azureml-fsspe:
• 1.0.0:
• Ubuntu:
• 3.9 and 3.10:

Describe the bug
When loading a parquet file from AzureML using azureml:// and pandas, python exits with segmentation fault. When using AzureMachineLearningFileSystem and fs.download() and then load the file once it is downloaded the file is loaded without any issues. However, I would like to load it without having to download it first.

This does not happen with every parquet file and I am unable to identify a difference for the cases it works and when it does not work.

To Reproduce
Steps to reproduce the behavior:
Not sure how to reproduce the file setup ...

Expected behavior
I would expect that loading the file using azureml:// and azureml-fsspec should work the same way as downloading the file and then loading it.