Kubelet View

The kubelet is the agent that runs on each node in a cluster. The kubelet view allows you to monitor a cluster.

Metric Description

You can view kubelet metrics in the following table.

**Table 1** Kubelet metrics
Metric	Unit	Description
Running Kubelets	N/A	The number of running kubelets in a cluster.
Running Pods	N/A	The number of running pods on the node where the kubelet resides.
Running Containers	N/A	The number of running containers on the node where the kubelet resides.
Actual Volumes	N/A	The actual number of volumes on the node where the kubelet resides.
Expected Volumes	N/A	The desired number of volumes on the node where the kubelet resides.
Configuration Errors	N/A	The number of incorrect Kubelet configurations on the node where the kubelet resides.
Operation Rate	N/A	The number of operations performed by the kubelet per second.
Operation Error Rate	N/A	The number of failed operations performed by the kubelet per second.
Operation Latency	second	The latencies of operations performed by the kubelet.
Pod Startup Rate	N/A	The number of pod startups performed by the kubelet per second.
Pod Startup Latency (P99)	second	The P99 latencies of pod startups performed by the kubelet.
Storage Operation Rate	N/A	The number of storage-related operations performed by the kubelet per second.
Storage Operation Error Rate	N/A	The number of failed storage-related operations performed by the kubelet per second.
Storage Operation Latency (P99)	second	The P99 latencies of storage-related operations performed by the kubelet.
Cgroup Manager Operation Rate	N/A	The number of destroy or update operations performed by the kubelet per second.
Cgroup Manager Operation Latency (P99)	second	The P99 latencies of destroy or update operations performed by the kubelet.
PLEG Relist Rate	N/A	The number of relisting operations in PLEG per second.
PLEG Relist Interval (P99)	second	The intervals between 99% of relisting operations in PLEG.
PLEG Relist Latency (P99)	second	The P99 latencies of relisting operations in PLEG.
RPC Rate	N/A	The number of RPC requests with each status code.
Request Latency (P99)	second	The P99 latencies of requests with each method.
Used Memory	byte	The memory used by the kubelet.
Used CPU	byte	The CPU used by the kubelet.
Goroutines	N/A	The number of goroutines.

Metric List

The following is the metric list of the kubelet view.

**Table 2** Metric description
Metric	Type	Description
storage_operation_errors_total	Counter	The number of errors in storage operations.
storage_operation_duration_seconds_count	Counter	The number of storage operations.
storage_operation_duration_seconds_bucket	Histogram	Duration for each storage operation.
kubelet_pod_start_duration_seconds_count	Counter	The number of pods that have been started.
kubelet_pod_start_duration_seconds_bucket	Histogram	Duration from the kubelet seeing a pod for the first time to the pod starting to run.
kubelet_runtime_operations_duration_seconds_bucket	Histogram	The time of every operation.
kubelet_runtime_operations_errors_total	Counter	The number of errors in operations at runtime level.
kubelet_node_config_error	Gauge	If a configuration-related error occurs on a node, the value of this metric is true (1). If there is no configuration-related error, the value is false (0).
volume_manager_total_volumes	Gauge	The number of volumes in Volume Manager.
kubelet_running_containers	Gauge	The number of running containers.
kubelet_running_pods	Gauge	The number of running pods.
kubelet_node_name	Gauge	Node name. The value is always 1.
kubelet_runtime_operations_total	Counter	The number of total runtime operations of each type.
kubelet_cgroup_manager_duration_seconds_count	Counter	The number of destruction and update operations.
kubelet_cgroup_manager_duration_seconds_bucket	Histogram	Duration for destruction and update operations.
kubelet_pleg_relist_duration_seconds_count	Counter	The number of relisting operations in PLEG.
kubelet_pleg_relist_interval_seconds_bucket	Histogram	The intervals between relisting operations in PLEG.
kubelet_pleg_relist_duration_seconds_bucket	Histogram	Duration for relisting pods in PLEG.
rest_client_requests_total	Counter	The total number of HTTP requests, partitioned by status code and method.
rest_client_request_duration_seconds_bucket	Histogram	The number of HTTP requests, partitioned by status code and method.
process_resident_memory_bytes	Gauge	Resident memory size in bytes.
process_cpu_seconds_total	Counter	Total user and system CPU time spent in seconds.
go_goroutines	Gauge	The number of goroutines.