An administrator is maintaining several Kubernetes clusters deployed through a Supervisor Namespace in a vSphere...

In VMware Cloud Foundation (VCF) 9.0 and vSphere Kubernetes Service (VKS) environments, the kubectl logs command is the essential tool for retrieving the standard output (STDOUT) and standard error (STDERR) streams from containers within a pod. When a microservice like an API gateway enters a CrashLoopBackOff state, it signifies that the container is starting, failing, and being restarted by the kubelet in an exponential back-off loop. To diagnose the root cause—such as a syntax error in a configuration file or a failed connection to a backend database—the administrator must inspect the application ' s internal execution logs.

The kubectl logs command is particularly powerful in this scenario when used with the -p (or --previous) flag. This flag allows the administrator to retrieve logs from the specific container instance that just crashed, rather than the one currently attempting to restart. While kubectl describe provides critical metadata about the pod ' s lifecycle events, resource constraints, and status, it does not capture the internal application stream. In VCF 9.0, where the vSphere Supervisor manages the lifecycle of these clusters, utilizing the standard Kubernetes toolset ensures that platform operators have a consistent and familiar troubleshooting experience. This logging mechanism is a core part of the observability stack in VKS, enabling rapid identification of application-level failures within the software-defined data center.