When using *.csv file format, a schema.ini file is automatically created. This file is used to “record” the data type in each column of the file and the number of columns. This is by design- a text file driver is used to read the *.csv, the first lines of data are read and the data type is determined. Different drivers are used to access different types of files - some drivers determine the data type internally.
It’s possible that the schema.ini files that were created on each setup are different. This can occur if you are making changes to how the file handling command in regard to the columns and data types. Try copying over the schema.ini file from the system it’s working on (or deleting the existing schema.ini file on the problematic system).