During the PAS Cloud auto-upgrade window on May 21st, several customer instances incurred approximately one hour of downtime. This was root caused to a backend system resource issue by the on-call engineer. The on-call engineering team manually intervened and completed auto-upgrades for the affected customer instances. A post-mortem was performed to diagnose the resourcing issue, and several areas of improvement were identified in PAS Cloud backend services. Implementation for these improvements is underway, and will prevent a recurrence of this same issue in the future.
Posted May 27, 2026 - 15:57 EDT
Resolved
Cloud Customers scheduled to be auto upgraded on Thursday May 21st may have experienced extended downtime of 1 to 4 hours. The team was able to complete the upgrades during the extended outage and service should now be restored to all customers. A RCA is being conducted to determine the cause of the extended upgrade period to prevent future outages.