I see I indeed got an alert at 3:51 UTC, roughly when you got this issue, but now it seems to be working fine again. I'll be monitoring the database all day but hopefully the service should go back to normal now.