Thanks to all for your input, we have solved the issue by building a docker for the process and bypassing the issue until the sys admins can sort it out :)