diff --git a/hack/.golint_failures b/hack/.golint_failures
index 2d9dbd53c6a..83dad22d368 100644
--- a/hack/.golint_failures
+++ b/hack/.golint_failures
@@ -129,7 +129,6 @@ pkg/kubelet/dockershim/network/kubenet
 pkg/kubelet/dockershim/network/testing
 pkg/kubelet/events
 pkg/kubelet/lifecycle
-pkg/kubelet/metrics
 pkg/kubelet/pluginmanager/pluginwatcher
 pkg/kubelet/pod/testing
 pkg/kubelet/preemption
diff --git a/pkg/kubelet/config/apiserver.go b/pkg/kubelet/config/apiserver.go
index a8afc6e345c..2f29ddd576f 100644
--- a/pkg/kubelet/config/apiserver.go
+++ b/pkg/kubelet/config/apiserver.go
@@ -14,7 +14,6 @@ See the License for the specific language governing permissions and
 limitations under the License.
 */
 
-// Reads the pod configuration from the Kubernetes apiserver.
 package config
 
 import (
diff --git a/pkg/kubelet/config/common.go b/pkg/kubelet/config/common.go
index 4e1f0f13009..cce17064529 100644
--- a/pkg/kubelet/config/common.go
+++ b/pkg/kubelet/config/common.go
@@ -14,7 +14,6 @@ See the License for the specific language governing permissions and
 limitations under the License.
 */
 
-// Common logic used by both http and file channels.
 package config
 
 import (
@@ -30,6 +29,7 @@ import (
 	utilyaml "k8s.io/apimachinery/pkg/util/yaml"
 	api "k8s.io/kubernetes/pkg/apis/core"
 	"k8s.io/kubernetes/pkg/apis/core/helper"
+
 	// TODO: remove this import if
 	// api.Registry.GroupOrDie(v1.GroupName).GroupVersion.String() is changed
 	// to "v1"?
@@ -110,6 +110,7 @@ func getSelfLink(name, namespace string) string {
 
 type defaultFunc func(pod *api.Pod) error
 
+// tryDecodeSinglePod takes data and tries to extract valid Pod config information from it.
 func tryDecodeSinglePod(data []byte, defaultFn defaultFunc) (parsed bool, pod *v1.Pod, err error) {
 	// JSON is valid YAML, so this should work for everything.
 	json, err := utilyaml.ToJSON(data)
diff --git a/pkg/kubelet/config/defaults.go b/pkg/kubelet/config/defaults.go
index 6c1e4ebf411..f2564e65d90 100644
--- a/pkg/kubelet/config/defaults.go
+++ b/pkg/kubelet/config/defaults.go
@@ -16,6 +16,7 @@ limitations under the License.
 
 package config
 
+// Defines sane defaults for the kubelet config.
 const (
 	DefaultKubeletPodsDirName                = "pods"
 	DefaultKubeletVolumesDirName             = "volumes"
diff --git a/pkg/kubelet/config/file.go b/pkg/kubelet/config/file.go
index ac07882b73c..dd3b38f7c4a 100644
--- a/pkg/kubelet/config/file.go
+++ b/pkg/kubelet/config/file.go
@@ -14,7 +14,6 @@ See the License for the specific language governing permissions and
 limitations under the License.
 */
 
-// Reads the pod configuration from file or a directory of files.
 package config
 
 import (
@@ -60,6 +59,7 @@ type sourceFile struct {
 	watchEvents    chan *watchEvent
 }
 
+// NewSourceFile watches a config file for changes.
 func NewSourceFile(path string, nodeName types.NodeName, period time.Duration, updates chan<- interface{}) {
 	// "github.com/sigma/go-inotify" requires a path without trailing "/"
 	path = strings.TrimRight(path, string(os.PathSeparator))
@@ -196,6 +196,7 @@ func (s *sourceFile) extractFromDir(name string) ([]*v1.Pod, error) {
 	return pods, nil
 }
 
+// extractFromFile parses a file for Pod configuration information.
 func (s *sourceFile) extractFromFile(filename string) (pod *v1.Pod, err error) {
 	klog.V(3).Infof("Reading config file %q", filename)
 	defer func() {
@@ -232,7 +233,7 @@ func (s *sourceFile) extractFromFile(filename string) (pod *v1.Pod, err error) {
 		return pod, nil
 	}
 
-	return pod, fmt.Errorf("%v: couldn't parse as pod(%v), please check config file.\n", filename, podErr)
+	return pod, fmt.Errorf("%v: couldn't parse as pod(%v), please check config file", filename, podErr)
 }
 
 func (s *sourceFile) replaceStore(pods ...*v1.Pod) (err error) {
diff --git a/pkg/kubelet/config/file_linux.go b/pkg/kubelet/config/file_linux.go
index b7d8712e325..ad730e47e34 100644
--- a/pkg/kubelet/config/file_linux.go
+++ b/pkg/kubelet/config/file_linux.go
@@ -16,7 +16,6 @@ See the License for the specific language governing permissions and
 limitations under the License.
 */
 
-// Reads the pod configuration from file or a directory of files.
 package config
 
 import (
@@ -50,17 +49,17 @@ func (e *retryableError) Error() string {
 
 func (s *sourceFile) startWatch() {
 	backOff := flowcontrol.NewBackOff(retryPeriod, maxRetryPeriod)
-	backOffId := "watch"
+	backOffID := "watch"
 
 	go wait.Forever(func() {
-		if backOff.IsInBackOffSinceUpdate(backOffId, time.Now()) {
+		if backOff.IsInBackOffSinceUpdate(backOffID, time.Now()) {
 			return
 		}
 
 		if err := s.doWatch(); err != nil {
 			klog.Errorf("Unable to read config path %q: %v", s.path, err)
 			if _, retryable := err.(*retryableError); !retryable {
-				backOff.Next(backOffId, time.Now())
+				backOff.Next(backOffID, time.Now())
 			}
 		}
 	}, retryPeriod)
@@ -130,11 +129,11 @@ func (s *sourceFile) produceWatchEvent(e *fsnotify.Event) error {
 func (s *sourceFile) consumeWatchEvent(e *watchEvent) error {
 	switch e.eventType {
 	case podAdd, podModify:
-		if pod, err := s.extractFromFile(e.fileName); err != nil {
+		pod, err := s.extractFromFile(e.fileName)
+		if err != nil {
 			return fmt.Errorf("can't process config file %q: %v", e.fileName, err)
-		} else {
-			return s.store.Add(pod)
 		}
+		return s.store.Add(pod)
 	case podDelete:
 		if objKey, keyExist := s.fileKeyMapping[e.fileName]; keyExist {
 			pod, podExist, err := s.store.GetByKey(objKey)
@@ -145,9 +144,8 @@ func (s *sourceFile) consumeWatchEvent(e *watchEvent) error {
 			} else {
 				if err = s.store.Delete(pod); err != nil {
 					return fmt.Errorf("failed to remove deleted pod from cache: %v", err)
-				} else {
-					delete(s.fileKeyMapping, e.fileName)
 				}
+				delete(s.fileKeyMapping, e.fileName)
 			}
 		}
 	}
diff --git a/pkg/kubelet/config/file_unsupported.go b/pkg/kubelet/config/file_unsupported.go
index d46b5f361de..e6f02de0fc9 100644
--- a/pkg/kubelet/config/file_unsupported.go
+++ b/pkg/kubelet/config/file_unsupported.go
@@ -16,7 +16,6 @@ See the License for the specific language governing permissions and
 limitations under the License.
 */
 
-// Reads the pod configuration from file or a directory of files.
 package config
 
 import (
diff --git a/pkg/kubelet/config/flags.go b/pkg/kubelet/config/flags.go
index bb5a97122f0..458901205db 100644
--- a/pkg/kubelet/config/flags.go
+++ b/pkg/kubelet/config/flags.go
@@ -23,9 +23,9 @@ import (
 	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
 )
 
+// ContainerRuntimeOptions defines options for the container runtime.
 type ContainerRuntimeOptions struct {
-
-	// General options.
+	// General Options.
 
 	// ContainerRuntime is the container runtime to use.
 	ContainerRuntime string
@@ -78,6 +78,7 @@ type ContainerRuntimeOptions struct {
 	CNICacheDir string
 }
 
+// AddFlags adds flags to the container runtime, according to ContainerRuntimeOptions.
 func (s *ContainerRuntimeOptions) AddFlags(fs *pflag.FlagSet) {
 	dockerOnlyWarning := "This docker-specific flag only works when container-runtime is set to docker."
 
diff --git a/pkg/kubelet/config/http.go b/pkg/kubelet/config/http.go
index 1dbef5f3d85..d0e3727e233 100644
--- a/pkg/kubelet/config/http.go
+++ b/pkg/kubelet/config/http.go
@@ -14,7 +14,6 @@ See the License for the specific language governing permissions and
 limitations under the License.
 */
 
-// Reads the pod configuration from an HTTP GET response.
 package config
 
 import (
@@ -43,6 +42,7 @@ type sourceURL struct {
 	client      *http.Client
 }
 
+// NewSourceURL specifies the URL where to read the Pod configuration from, then watches it for changes.
 func NewSourceURL(url string, header http.Header, nodeName types.NodeName, period time.Duration, updates chan<- interface{}) {
 	config := &sourceURL{
 		url:      url,
diff --git a/pkg/kubelet/configmap/configmap_manager.go b/pkg/kubelet/configmap/configmap_manager.go
index 58c075693b1..132f60c18f2 100644
--- a/pkg/kubelet/configmap/configmap_manager.go
+++ b/pkg/kubelet/configmap/configmap_manager.go
@@ -20,7 +20,7 @@ import (
 	"fmt"
 	"time"
 
-	"k8s.io/api/core/v1"
+	v1 "k8s.io/api/core/v1"
 	clientset "k8s.io/client-go/kubernetes"
 	podutil "k8s.io/kubernetes/pkg/api/v1/pod"
 	corev1 "k8s.io/kubernetes/pkg/apis/core/v1"
diff --git a/pkg/kubelet/configmap/fake_manager.go b/pkg/kubelet/configmap/fake_manager.go
index 959ab6b693f..cff8f1fe1c9 100644
--- a/pkg/kubelet/configmap/fake_manager.go
+++ b/pkg/kubelet/configmap/fake_manager.go
@@ -16,9 +16,7 @@ limitations under the License.
 
 package configmap
 
-import (
-	"k8s.io/api/core/v1"
-)
+import v1 "k8s.io/api/core/v1"
 
 // fakeManager implements Manager interface for testing purposes.
 // simple operations to apiserver.
diff --git a/pkg/kubelet/metrics/metrics.go b/pkg/kubelet/metrics/metrics.go
index e8b83f70261..83f175b2461 100644
--- a/pkg/kubelet/metrics/metrics.go
+++ b/pkg/kubelet/metrics/metrics.go
@@ -30,6 +30,7 @@ import (
 	kubecontainer "k8s.io/kubernetes/pkg/kubelet/container"
 )
 
+// This const block defines the metric names for the kubelet metrics.
 const (
 	KubeletSubsystem                     = "kubelet"
 	NodeNameKey                          = "node_name"
@@ -85,6 +86,7 @@ const (
 )
 
 var (
+	// NodeName is a Gauge that tracks the ode's name. The count is always 1.
 	NodeName = prometheus.NewGaugeVec(
 		prometheus.GaugeOpts{
 			Subsystem: KubeletSubsystem,
@@ -93,6 +95,7 @@ var (
 		},
 		[]string{NodeLabelKey},
 	)
+	// ContainersPerPodCount is a Counter that tracks the number of containers per pod.
 	ContainersPerPodCount = prometheus.NewHistogram(
 		prometheus.HistogramOpts{
 			Subsystem: KubeletSubsystem,
@@ -101,6 +104,8 @@ var (
 			Buckets:   prometheus.DefBuckets,
 		},
 	)
+	// PodWorkerDuration is a Histogram that tracks the duration (in seconds) in takes to sync a single pod.
+	// Broken down by the operation type.
 	PodWorkerDuration = prometheus.NewHistogramVec(
 		prometheus.HistogramOpts{
 			Subsystem: KubeletSubsystem,
@@ -110,6 +115,7 @@ var (
 		},
 		[]string{"operation_type"},
 	)
+	// PodStartDuration is a Histogram that tracks the duration (in seconds) it takes for a single pod to go from pending to running.
 	PodStartDuration = prometheus.NewHistogram(
 		prometheus.HistogramOpts{
 			Subsystem: KubeletSubsystem,
@@ -118,6 +124,8 @@ var (
 			Buckets:   prometheus.DefBuckets,
 		},
 	)
+	// CgroupManagerDuration is a Histogram that tracks the duration (in seconds) it takes for cgroup manager operations to complete.
+	// Broken down by method.
 	CgroupManagerDuration = prometheus.NewHistogramVec(
 		prometheus.HistogramOpts{
 			Subsystem: KubeletSubsystem,
@@ -127,6 +135,7 @@ var (
 		},
 		[]string{"operation_type"},
 	)
+	// PodWorkerStartDuration is a Histogram that tracks the duration (in seconds) it takes from seeing a pod to starting a worker.
 	PodWorkerStartDuration = prometheus.NewHistogram(
 		prometheus.HistogramOpts{
 			Subsystem: KubeletSubsystem,
@@ -135,6 +144,8 @@ var (
 			Buckets:   prometheus.DefBuckets,
 		},
 	)
+	// PLEGRelistDuration is a Histogram that tracks the duration (in seconds) it takes for relisting pods in the Kubelet's
+	// Pod Lifecycle Event Generator (PLEG).
 	PLEGRelistDuration = prometheus.NewHistogram(
 		prometheus.HistogramOpts{
 			Subsystem: KubeletSubsystem,
@@ -143,6 +154,8 @@ var (
 			Buckets:   prometheus.DefBuckets,
 		},
 	)
+	// PLEGDiscardEvents is a Histogram that tracks the duration (in seconds) it takes for discarding events in the Kubelet's
+	// Pod Lifecycle Event Generator (PLEG).
 	PLEGDiscardEvents = prometheus.NewCounterVec(
 		prometheus.CounterOpts{
 			Subsystem: KubeletSubsystem,
@@ -151,6 +164,8 @@ var (
 		},
 		[]string{},
 	)
+	// PLEGRelistInterval is a Histogram that tracks the intervals (in seconds) between relisting in the Kubelet's
+	// Pod Lifecycle Event Generator (PLEG).
 	PLEGRelistInterval = prometheus.NewHistogram(
 		prometheus.HistogramOpts{
 			Subsystem: KubeletSubsystem,
@@ -159,7 +174,8 @@ var (
 			Buckets:   prometheus.DefBuckets,
 		},
 	)
-	// Metrics of remote runtime operations.
+	// RuntimeOperations is a Counter that tracks the cumulative number of remote runtime operations.
+	// Broken down by operation type.
 	RuntimeOperations = prometheus.NewCounterVec(
 		prometheus.CounterOpts{
 			Subsystem: KubeletSubsystem,
@@ -168,6 +184,8 @@ var (
 		},
 		[]string{"operation_type"},
 	)
+	// RuntimeOperationsDuration is a Histogram that tracks the duration (in seconds) for remote runtime operations to complete.
+	// Broken down by operation type.
 	RuntimeOperationsDuration = prometheus.NewHistogramVec(
 		prometheus.HistogramOpts{
 			Subsystem: KubeletSubsystem,
@@ -177,6 +195,8 @@ var (
 		},
 		[]string{"operation_type"},
 	)
+	// RuntimeOperationsErrors is a Counter that tracks the cumulative number of remote runtime operations errors.
+	// Broken down by operation type.
 	RuntimeOperationsErrors = prometheus.NewCounterVec(
 		prometheus.CounterOpts{
 			Subsystem: KubeletSubsystem,
@@ -185,6 +205,8 @@ var (
 		},
 		[]string{"operation_type"},
 	)
+	// EvictionStatsAge is a Histogram that tracks the time (in seconds) between when stats are collected and when a pod is evicted
+	// based on those stats. Broken down by eviction signal.
 	EvictionStatsAge = prometheus.NewHistogramVec(
 		prometheus.HistogramOpts{
 			Subsystem: KubeletSubsystem,
@@ -194,6 +216,8 @@ var (
 		},
 		[]string{"eviction_signal"},
 	)
+	// DevicePluginRegistrationCount is a Counter that tracks the cumulative number of device plugin registrations.
+	// Broken down by resource name.
 	DevicePluginRegistrationCount = prometheus.NewCounterVec(
 		prometheus.CounterOpts{
 			Subsystem: KubeletSubsystem,
@@ -202,6 +226,8 @@ var (
 		},
 		[]string{"resource_name"},
 	)
+	// DevicePluginAllocationDuration is a Histogram that tracks the duration (in seconds) to serve a device plugin allocation request.
+	// Broken down by resource name.
 	DevicePluginAllocationDuration = prometheus.NewHistogramVec(
 		prometheus.HistogramOpts{
 			Subsystem: KubeletSubsystem,
@@ -211,7 +237,8 @@ var (
 		},
 		[]string{"resource_name"},
 	)
-
+	// DeprecatedPodWorkerLatency is a Summary that tracks the latency (in microseconds) to sync a single pod.
+	// Broken down by operation type. This metric is deprecated.
 	DeprecatedPodWorkerLatency = prometheus.NewSummaryVec(
 		prometheus.SummaryOpts{
 			Subsystem: KubeletSubsystem,
@@ -220,6 +247,8 @@ var (
 		},
 		[]string{"operation_type"},
 	)
+	// DeprecatedPodStartLatency is a Summary that tracks the latency (in microseconds) for a single pod to go from pending to running.
+	// This metric is deprecated.
 	DeprecatedPodStartLatency = prometheus.NewSummary(
 		prometheus.SummaryOpts{
 			Subsystem: KubeletSubsystem,
@@ -227,6 +256,8 @@ var (
 			Help:      "(Deprecated) Latency in microseconds for a single pod to go from pending to running.",
 		},
 	)
+	// DeprecatedCgroupManagerLatency is a Summary that tracks the latency (in microseconds) for cgroup manager operations to complete.
+	// Broken down by operation type. This metric is deprecated.
 	DeprecatedCgroupManagerLatency = prometheus.NewSummaryVec(
 		prometheus.SummaryOpts{
 			Subsystem: KubeletSubsystem,
@@ -235,6 +266,8 @@ var (
 		},
 		[]string{"operation_type"},
 	)
+	// DeprecatedPodWorkerStartLatency is a Summary that tracks the latency (in microseconds) from seeing a pod to starting a worker.
+	// This metric is deprecated.
 	DeprecatedPodWorkerStartLatency = prometheus.NewSummary(
 		prometheus.SummaryOpts{
 			Subsystem: KubeletSubsystem,
@@ -242,6 +275,8 @@ var (
 			Help:      "(Deprecated) Latency in microseconds from seeing a pod to starting a worker.",
 		},
 	)
+	// DeprecatedPLEGRelistLatency is a Summary that tracks the latency (in microseconds) for relisting pods in PLEG.
+	// This metric is deprecated.
 	DeprecatedPLEGRelistLatency = prometheus.NewSummary(
 		prometheus.SummaryOpts{
 			Subsystem: KubeletSubsystem,
@@ -249,6 +284,8 @@ var (
 			Help:      "(Deprecated) Latency in microseconds for relisting pods in PLEG.",
 		},
 	)
+	// DeprecatedPLEGRelistInterval is a Summary that tracks the interval (in microseconds) between relistings in PLEG.
+	// This metric is deprecated.
 	DeprecatedPLEGRelistInterval = prometheus.NewSummary(
 		prometheus.SummaryOpts{
 			Subsystem: KubeletSubsystem,
@@ -256,6 +293,8 @@ var (
 			Help:      "(Deprecated) Interval in microseconds between relisting in PLEG.",
 		},
 	)
+	// DeprecatedRuntimeOperations is a Counter that tracks the cumulative number of remote runtime operations.
+	// Broken down by operation type. This metric is deprecated.
 	DeprecatedRuntimeOperations = prometheus.NewCounterVec(
 		prometheus.CounterOpts{
 			Subsystem: KubeletSubsystem,
@@ -264,6 +303,8 @@ var (
 		},
 		[]string{"operation_type"},
 	)
+	// DeprecatedRuntimeOperationsLatency is a Summary that tracks the latency (in microseconds) of remote runtime operations
+	// to complete. Broken down by operation type. This metric is deprecated.
 	DeprecatedRuntimeOperationsLatency = prometheus.NewSummaryVec(
 		prometheus.SummaryOpts{
 			Subsystem: KubeletSubsystem,
@@ -272,6 +313,8 @@ var (
 		},
 		[]string{"operation_type"},
 	)
+	// DeprecatedRuntimeOperationsErrors is a Counter that tracks the cumulative number of remote runtime operation errors.
+	// Broken down by operation type. This metric is deprecated.
 	DeprecatedRuntimeOperationsErrors = prometheus.NewCounterVec(
 		prometheus.CounterOpts{
 			Subsystem: KubeletSubsystem,
@@ -280,6 +323,8 @@ var (
 		},
 		[]string{"operation_type"},
 	)
+	// DeprecatedEvictionStatsAge is a Summary that tracks the time (in microseconds) between when stats are collected and when a pod
+	// is evicted based on those stats. Broken down by eviction signal. This metric is deprecated.
 	DeprecatedEvictionStatsAge = prometheus.NewSummaryVec(
 		prometheus.SummaryOpts{
 			Subsystem: KubeletSubsystem,
@@ -288,6 +333,8 @@ var (
 		},
 		[]string{"eviction_signal"},
 	)
+	// DeprecatedDevicePluginRegistrationCount is a Counter that tracks the cumulative number of device plugin registrations.
+	// Broken down by resource name. This metric is deprecated.
 	DeprecatedDevicePluginRegistrationCount = prometheus.NewCounterVec(
 		prometheus.CounterOpts{
 			Subsystem: KubeletSubsystem,
@@ -296,6 +343,8 @@ var (
 		},
 		[]string{"resource_name"},
 	)
+	// DeprecatedDevicePluginAllocationLatency is a Summary that tracks the latncy (in microseconds) for serving device plugin allocation requests.
+	// Broken down by resource name. This metric is deprecated.
 	DeprecatedDevicePluginAllocationLatency = prometheus.NewSummaryVec(
 		prometheus.SummaryOpts{
 			Subsystem: KubeletSubsystem,
@@ -307,6 +356,7 @@ var (
 
 	// Metrics for node config
 
+	// AssignedConfig is a Gauge that is set 1 if the Kubelet has a NodeConfig assigned.
 	AssignedConfig = prometheus.NewGaugeVec(
 		prometheus.GaugeOpts{
 			Subsystem: KubeletSubsystem,
@@ -315,6 +365,7 @@ var (
 		},
 		[]string{ConfigSourceLabelKey, ConfigUIDLabelKey, ConfigResourceVersionLabelKey, KubeletConfigKeyLabelKey},
 	)
+	// ActiveConfig is a Gauge that is set to 1 if the Kubelet has an active NodeConfig.
 	ActiveConfig = prometheus.NewGaugeVec(
 		prometheus.GaugeOpts{
 			Subsystem: KubeletSubsystem,
@@ -323,6 +374,8 @@ var (
 		},
 		[]string{ConfigSourceLabelKey, ConfigUIDLabelKey, ConfigResourceVersionLabelKey, KubeletConfigKeyLabelKey},
 	)
+	// LastKnownGoodConfig is a Gauge that is set to 1 if the Kubelet has a NodeConfig it can fall back to if there
+	// are certain errors.
 	LastKnownGoodConfig = prometheus.NewGaugeVec(
 		prometheus.GaugeOpts{
 			Subsystem: KubeletSubsystem,
@@ -331,6 +384,7 @@ var (
 		},
 		[]string{ConfigSourceLabelKey, ConfigUIDLabelKey, ConfigResourceVersionLabelKey, KubeletConfigKeyLabelKey},
 	)
+	// ConfigError is a Gauge that is set to 1 if the node is experiencing a configuration-related error.
 	ConfigError = prometheus.NewGauge(
 		prometheus.GaugeOpts{
 			Subsystem: KubeletSubsystem,
@@ -338,6 +392,8 @@ var (
 			Help:      "This metric is true (1) if the node is experiencing a configuration-related error, false (0) otherwise.",
 		},
 	)
+	// RunPodSandboxDuration is a Histogram that tracks the duration (in seconds) it takes to run Pod Sandbox operations.
+	// Broken down by RuntimeClass.
 	RunPodSandboxDuration = prometheus.NewHistogramVec(
 		prometheus.HistogramOpts{
 			Subsystem: KubeletSubsystem,
@@ -348,6 +404,8 @@ var (
 		},
 		[]string{"runtime_handler"},
 	)
+	// RunPodSandboxErrors is a Counter that tracks the cumulative number of Pod Sandbox operations errors.
+	// Broken down by RuntimeClass.
 	RunPodSandboxErrors = prometheus.NewCounterVec(
 		prometheus.CounterOpts{
 			Subsystem: KubeletSubsystem,
@@ -360,7 +418,7 @@ var (
 
 var registerMetrics sync.Once
 
-// Register all metrics.
+// Register registers all metrics.
 func Register(containerCache kubecontainer.RuntimeCache, collectors ...prometheus.Collector) {
 	// Register the metrics.
 	registerMetrics.Do(func() {
@@ -404,7 +462,7 @@ func Register(containerCache kubecontainer.RuntimeCache, collectors ...prometheu
 	})
 }
 
-// Gets the time since the specified start in microseconds.
+// SinceInMicroseconds gets the time since the specified start in microseconds.
 func SinceInMicroseconds(start time.Time) float64 {
 	return float64(time.Since(start).Nanoseconds() / time.Microsecond.Nanoseconds())
 }
@@ -438,11 +496,15 @@ var (
 		nil, nil)
 )
 
+// Describe implements Prometheus' Describe method from the Collector interface. It sends all
+// available descriptions to the provided channel and retunrs once the last description has been sent.
 func (pc *podAndContainerCollector) Describe(ch chan<- *prometheus.Desc) {
 	ch <- runningPodCountDesc
 	ch <- runningContainerCountDesc
 }
 
+// Collect implements Prometheus' Collect method from the Collector interface. It's called by the Prometheus
+// registry when collecting metrics.
 func (pc *podAndContainerCollector) Collect(ch chan<- prometheus.Metric) {
 	runningPods, err := pc.containerCache.GetPods()
 	if err != nil {
@@ -488,8 +550,10 @@ func configLabels(source *corev1.NodeConfigSource) (map[string]string, error) {
 }
 
 // track labels across metric updates, so we can delete old label sets and prevent leaks
-var assignedConfigLabels map[string]string = map[string]string{}
+var assignedConfigLabels map[string]string
 
+// SetAssignedConfig tracks labels according to the assigned NodeConfig. It also tracks labels
+// across metric updates so old labels can be safely deleted.
 func SetAssignedConfig(source *corev1.NodeConfigSource) error {
 	// compute the timeseries labels from the source
 	labels, err := configLabels(source)
@@ -508,8 +572,10 @@ func SetAssignedConfig(source *corev1.NodeConfigSource) error {
 }
 
 // track labels across metric updates, so we can delete old label sets and prevent leaks
-var activeConfigLabels map[string]string = map[string]string{}
+var activeConfigLabels map[string]string
 
+// SetActiveConfig tracks labels according to the NodeConfig that is currently used by the Kubelet.
+// It also tracks labels across metric updates so old labels can be safely deleted.
 func SetActiveConfig(source *corev1.NodeConfigSource) error {
 	// compute the timeseries labels from the source
 	labels, err := configLabels(source)
@@ -528,8 +594,10 @@ func SetActiveConfig(source *corev1.NodeConfigSource) error {
 }
 
 // track labels across metric updates, so we can delete old label sets and prevent leaks
-var lastKnownGoodConfigLabels map[string]string = map[string]string{}
+var lastKnownGoodConfigLabels map[string]string
 
+// SetLastKnownGoodConfig tracks labels according to the NodeConfig that was successfully applied last.
+// It also tracks labels across metric updates so old labels can be safely deleted.
 func SetLastKnownGoodConfig(source *corev1.NodeConfigSource) error {
 	// compute the timeseries labels from the source
 	labels, err := configLabels(source)
@@ -547,6 +615,7 @@ func SetLastKnownGoodConfig(source *corev1.NodeConfigSource) error {
 	return nil
 }
 
+// SetConfigError sets a the ConfigError metric to 1 in case any errors were encountered.
 func SetConfigError(err bool) {
 	if err {
 		ConfigError.Set(1)
@@ -555,6 +624,7 @@ func SetConfigError(err bool) {
 	}
 }
 
+// SetNodeName sets the NodeName Gauge to 1.
 func SetNodeName(name types.NodeName) {
 	NodeName.WithLabelValues(string(name)).Set(1)
 }
diff --git a/pkg/kubelet/pod/mirror_client.go b/pkg/kubelet/pod/mirror_client.go
index 79d8cecd01c..afd60f5df13 100644
--- a/pkg/kubelet/pod/mirror_client.go
+++ b/pkg/kubelet/pod/mirror_client.go
@@ -105,13 +105,13 @@ func (mc *basicMirrorClient) DeleteMirrorPod(podFullName string, uid *types.UID)
 	return true, nil
 }
 
-// IsStaticPod returns true if the pod is a static pod.
+// IsStaticPod returns true if the passed Pod is static.
 func IsStaticPod(pod *v1.Pod) bool {
 	source, err := kubetypes.GetPodSource(pod)
 	return err == nil && source != kubetypes.ApiserverSource
 }
 
-// IsMirrorPod returns true if the pod is a mirror pod.
+// IsMirrorPod returns true if the passed Pod is a Mirror Pod.
 func IsMirrorPod(pod *v1.Pod) bool {
 	_, ok := pod.Annotations[kubetypes.ConfigMirrorAnnotationKey]
 	return ok