An incident with our compute provider specific to NYC systems, starting at Mar 05, 2018 - 03:57 UTC and being resolved at Mar 05, 2018 - 07:15 UTC prevent consistent capture from workers in that region. At this time, the incident seems fully resolved.