When importing from a CSV file, if the encoding of the file does not match the platform default encoding the import fails to import issues containing "invalid" characters.

Refer to JSP-53202

The problem causes the issue not to be imported when MySQL is the target database, but other databases import the issue, but with the incorrect character substituted for the invalid characters. If the platform uses UTF-8 encoding then the substituted character is the universal UTF-8 substitute character \xEF\xBF\xBD.

This might typically occur when the csv file is produced on a Windows machine and encoded as ISO-8859-1 and JIRA is running on a linux platform with its default encoding set to UTF-8.

A possible solution might be to amend the UI to allow the user to optionally specify the input file's encoding, so that the input is handled correctly.