Caused T146581: celery log level is INFO causing disruption on ORES service
We should have some way of alerting (icinga?) when disk space is running low on the scb* machines and the ORES labs machines.
This task is done when we have such alerts in place.