Upgrade complete. Performance may be degraded for a few hours as internal caches and statistics rebuild. We will continue to monitor performance during and after the maintenance window.

Standby has now caught up. Continuing upgrade procedure. Expect up to 30 minutes of downtime while we perform the upgrade.

Provisioning a HA standby has been taking longer than expected, so we are extending the maintenance window. The API and Dashboard are currently up and available, and we only recorded 23 seconds of downtime during the first portion of the upgrade procedure. We'll publish another update whenever the standby has caught up and the upgrade procedure resumes.

We're seeing a volume increase this year across multiple large accounts, and we need to upgrade our database infrastructure to support this increase moving forward. We will be taking an outage tonight to facilitate the upgrade and changeover. We will attempt to minimize downtime, but expect at least 5 minutes of downtime during the changeover. Request timing may increase or become sporadic throughout the day as we prepare the database for the upgrade procedure. Sorry for the inconvenience.