After an exhaustive investigation, we have determined that this outage of the API has been due to the lack of maintenance of the infrastructure that hosts this service.
Going forward we will do monthly rather than yearly reboots to our infrastructure.