Earlier today, we ran into issues with scaling our systems in the Mumbai region, primarily because of an edge case identified in AWS's instance allocation. This resulted in a high error rate for customers using the Mumbai processing region a couple of times.
After identifying the issue, which lasted for about 16 minutes around 11:50AM IST and 8 minutes around 09:45AM IST, our team made some changes to get around this limitation and restore functionality in this region.
The service is completely functional now with no degradation in response times either.
Our team is also working on a long-term fix so that the corner case encountered today does not impact services in the future. We remain committed to ensuring an uptime as close to 100% as possible.