kubectl set-context [CONTEXT_NAME] --namespace=[NAMESPACE] If CONTEXT_NAME is omitted, the current context (as indicated by current-context in kubeconfig) is modified. The command modifies a YAML structure inside the user's $HOME/.kube/config . Consider an initial context entry:
The command kubectl set-context --namespace allows a user to bind a default namespace to a specific context —a named cluster-user pair stored in the kubeconfig file. Once set, all subsequent kubectl operations that support namespace scoping implicitly execute within that default namespace unless explicitly overridden. kubectl set-context namespace
# Set once kubectl set-context --namespace=payments kubectl get pods kubectl get services kubectl get deployments Override for a single command kubectl get pods --namespace=audit # operates in 'audit' 4.2 Interaction with --all-namespaces The --all-namespaces flag (or -A ) explicitly overrides any namespace setting—including the context default—and operates across all namespaces. This is consistent with the precedence rule: the imperative flag takes highest priority. 4.3 Namespace Existence and Validation Crucially, kubectl set-context --namespace does not verify that the specified namespace exists in the target cluster. The validation occurs at the moment of a resource operation (e.g., get , create ). If the namespace does not exist, the API server returns: Once set, all subsequent kubectl operations that support
kubectl set-context prod-ops --namespace=monitoring The context becomes: 4.3 Namespace Existence and Validation Crucially
contexts: - name: prod-ops context: cluster: prod-eks user: ops-user After executing: