Our operations team was notified at 9:49AM about high disk io on our production database. We determined that this was caused by a task activated during a regularly scheduled deploy. Inserts into the database were delayed during this 40 minute period.