All our services are at 99.25% for the past 15+ months.
We hit with 4xx/5xx error rates at ~18% along with increased latency up to 40 secs for our search queries on Oct 20 2025. It lasted for about 172 minutes. The increased error rates occurred due to our automation’s inability to flex up new nodes for handling additional re-tries and regular traffic. Now, we have the automation ability to detect multi-service AWS outages and cut over the traffic to the US-West region in under ~45 minutes.