Previous incidents

Post Mortem: We got a huge influx of requests that overwhelmed the API. Ee've scalated the services and are actively working on mitigating the issue in future occasions.

Over the weekend, during a period of unusually high load, our API experienced performance degradation caused by a database misconfiguration. The configuration inadvertently capped read/write operations and resulted in CPU resource constraints, creating a bottleneck under peak traffic conditions.

Following a thorough investigation, we optimized the database configuration to address these limitations. This resolution has led to a measurable improvement in API latency of up to 20%.

To mitigate...

We experienced a DNS misconfiguration that caused the API to return a 302 redirect, depending on the region. Our internal monitoring service, based in a different region, did not detect the issue immediately. We rolled out monitoring services across all regions to ensure earlier detection in the future.