Title is misleading. The deletion of the production database was last week. Today, it's down in a single region because of multiple redundant power sources failed.
SFO2 is offline.
Cloud API panel was having issues before that and affects all regions.