We've had an incident between 08:23 PT / 15:23 UTC and 8:35 PT / 15:35 UTC. We have fully recovered.
A fix has been implemented and we are monitoring the results.
We are currently investigating this issue.