Summary
We saw a sharp rise in inbound requests to the Payway SSO service. To protect availability we first scaled out infrastructure to absorb the burst, assuming an organic traffic spike. Further investigation showed the load was amplified by a latent performance issue that degraded throughput under pressure — a compounding effect rather than traffic alone. The issue has been fixed and the service is running steadily.
Timeline / Response
Root cause
Under elevated load, a performance issue caused throughput to degrade, and that degradation fed back on itself — compounding the impact and making the surge appear larger
than the underlying traffic warranted.
Resolution
We identified and fixed the underlying performance issue. Request rates and response times have returned to normal, and we confirm the service is now stable.