A downstream resource used by Kalepa's platform had performance issues which in turn degraded latency and performance of several dependent services. Once alternate capacity was provided for the degraded resource, caches had to be re-populated which led a to a delay in recovery. Resources are now over-provisioned for the compromised service and Kalepa is analyzing other dependencies and adding further instrumentation and resiliency around the resource.

We are seeing elevated error rates and increased latency across Kalepa platform. We are investigating the root cause.