Summary: Between 16:55 and 18:30 UTC on July 13 (~95 minutes), our authentication service experienced elevated latency that degraded features relying on real-time user-account lookups affecting our console service.
Impact: The majority of requests (roughly 90–95%) continued to complete successfully throughout, but a portion were slow or timed out. Affected users may have seen usernames or profile details slow to load, slower sign-in, or occasional errors on community interactions. This was a performance/availability issue only - no customer or end-user data was lost, exposed, or corrupted.
Cause & resolution: An internal service change unexpectedly increased load on our authentication service beyond its provisioned capacity. We resolved the incident by rolling the change back, after which service returned to normal.
Next steps: We're adding caching, capacity, and staged-rollout safeguards to prevent this from recurring. We apologize for the disruption.