Description
The problem runs on fleet-cast. Simulate a fault where a Kubernetes operator misconfigures replicas to an extremely high number (overload) in the TiDB cluster. The TiDBCluster custom resource is configured with an excessive number of replicas (100,000), overwhelming the cluster and causing only a few pods to be created successfully. Inject the fault by modifying the operator configuration to set replica count to an unreasonably high value.
Tags
Cloud Management System Failure, fleet-cast