Opened on Saturday 3rd February 2018, last updated
Resolved — We identified and made some database updates. They have successfully addressed the service disruption from the background maintenance tasks.
Monitoring — We have modified the maintenance task to space out the requests to the talk service. Accordingly the Talk service is back up, however it may still suffer some degraded performance during this period.
Monitoring — We've identified the cause as being a background maintenance task which is overwhelming the Talk API. We've added additional capacity to the Talk API and its database, and Talk is now back online. We're monitoring to ensure that it remains stable.