From c51a8734b12cffd20d366bf1b329ded7a470dce8 Mon Sep 17 00:00:00 2001 From: Richa Banker Date: Fri, 12 Sep 2025 10:24:32 -0700 Subject: [PATCH] Update documented metrics list --- .../resourceclaim/metrics/metrics.go | 3 +- pkg/kubelet/metrics/metrics.go | 3 +- .../documentation/documentation-list.yaml | 2012 ++++++++++------- .../documentation/documentation.md | 458 +++- 4 files changed, 1510 insertions(+), 966 deletions(-) diff --git a/pkg/controller/resourceclaim/metrics/metrics.go b/pkg/controller/resourceclaim/metrics/metrics.go index f9fe6f78fa7..b5a926ff3ea 100644 --- a/pkg/controller/resourceclaim/metrics/metrics.go +++ b/pkg/controller/resourceclaim/metrics/metrics.go @@ -42,7 +42,8 @@ var ( // NumResourceClaimsDesc tracks the number of ResourceClaims, // categorized by their allocation status and admin access. - NumResourceClaimsDesc = metrics.NewDesc(ResourceClaimSubsystem+"_resource_claims", + NumResourceClaimsDesc = metrics.NewDesc( + metrics.BuildFQName("", ResourceClaimSubsystem, "resource_claims"), "Number of ResourceClaims, categorized by allocation status and admin access", []string{"allocated", "admin_access"}, nil, metrics.ALPHA, "") diff --git a/pkg/kubelet/metrics/metrics.go b/pkg/kubelet/metrics/metrics.go index d960989d2bd..5c860d61384 100644 --- a/pkg/kubelet/metrics/metrics.go +++ b/pkg/kubelet/metrics/metrics.go @@ -1075,7 +1075,8 @@ var ( []string{"driver_name", "method_name", "grpc_status_code"}, ) - DRAResourceClaimsInUseDesc = metrics.NewDesc(DRASubsystem+"_resource_claims_in_use", + DRAResourceClaimsInUseDesc = metrics.NewDesc( + metrics.BuildFQName("", DRASubsystem, "resource_claims_in_use"), "The number of ResourceClaims that are currently in use on the node, by driver name (driver_name label value) and across all drivers (special value for driver_name). Note that the sum of all by-driver counts is not the total number of in-use ResourceClaims because the same ResourceClaim might use devices from different drivers. Instead, use the count for the driver_name.", []string{"driver_name"}, nil, diff --git a/test/instrumentation/documentation/documentation-list.yaml b/test/instrumentation/documentation/documentation-list.yaml index b94b377b3a3..52b351d5b0c 100644 --- a/test/instrumentation/documentation/documentation-list.yaml +++ b/test/instrumentation/documentation/documentation-list.yaml @@ -89,6 +89,30 @@ - 128 - 256 - 512 +- name: pod_deletion_duration_seconds + subsystem: device_taint_eviction_controller + help: Latency, in seconds, between the time when a device taint effect has been + activated and a Pod's deletion via DeviceTaintEvictionController. + type: Histogram + stabilityLevel: ALPHA + buckets: + - 0.005 + - 0.025 + - 0.1 + - 0.5 + - 1 + - 2.5 + - 10 + - 30 + - 60 + - 120 + - 180 + - 240 +- name: pod_deletions_total + subsystem: device_taint_eviction_controller + help: Total number of Pods deleted by DeviceTaintEvictionController since its start. + type: Counter + stabilityLevel: ALPHA - name: addresses_skipped_per_sync subsystem: endpoint_slice_mirroring_controller help: Number of addresses skipped on each Endpoints sync due to being invalid or @@ -476,26 +500,22 @@ - 2 - 4 - 8 -- name: allocated_resource_claims +- name: creates_total subsystem: resourceclaim_controller - help: Number of allocated ResourceClaims - type: Gauge - stabilityLevel: ALPHA -- name: create_attempts_total - subsystem: resourceclaim_controller - help: Number of ResourceClaims creation requests + help: Number of ResourceClaims creation requests, categorized by creation status + and admin access type: Counter stabilityLevel: ALPHA -- name: create_failures_total - subsystem: resourceclaim_controller - help: Number of ResourceClaims creation request failures - type: Counter - stabilityLevel: ALPHA -- name: resource_claims - subsystem: resourceclaim_controller - help: Number of ResourceClaims - type: Gauge + labels: + - admin_access + - status +- name: resourceclaim_controller_resource_claims + help: Number of ResourceClaims, categorized by allocation status and admin access + type: Custom stabilityLevel: ALPHA + labels: + - allocated + - admin_access - name: job_pods_finished_total subsystem: job_controller help: The number of finished Pods that are fully tracked @@ -579,6 +599,13 @@ help: Number of PersistentVolumeClaim creation requests type: Counter stabilityLevel: ALPHA +- name: kubelet_credential_provider_config_info + help: Information about the last applied credential provider configuration with + hash as label + type: Custom + stabilityLevel: ALPHA + labels: + - hash - name: credential_provider_plugin_duration subsystem: kubelet help: Duration of execution in seconds for credential provider plugin @@ -598,7 +625,7 @@ - 2.5 - 5 - 10 -- name: credential_provider_plugin_errors +- name: credential_provider_plugin_errors_total subsystem: kubelet help: Number of errors from credential provider plugin type: Counter @@ -755,6 +782,15 @@ help: Counter of certificate renewal errors. type: Counter stabilityLevel: ALPHA +- name: container_swap_limit_bytes + help: Current amount of the container swap limit in bytes. Reported only on non-windows + systems + type: Custom + stabilityLevel: ALPHA + labels: + - container + - pod + - namespace - name: container_swap_usage_bytes help: Current amount of the container swap usage in bytes. Reported only on non-windows systems @@ -817,16 +853,16 @@ - 16.995624819678714 - 26.07345379475354 - 39.99999999999997 -- name: force_cleaned_failed_volume_operation_errors_total - help: The number of volumes that failed force cleanup after their reconstruction - failed during kubelet startup. - type: Counter - stabilityLevel: ALPHA -- name: force_cleaned_failed_volume_operations_total - help: The number of volumes that were force cleaned after their reconstruction failed - during kubelet startup. This includes both successful and failed cleanups. - type: Counter +- name: dra_resource_claims_in_use + help: The number of ResourceClaims that are currently in use on the node, by driver + name (driver_name label value) and across all drivers (special value for + driver_name). Note that the sum of all by-driver counts is not the total number + of in-use ResourceClaims because the same ResourceClaim might use devices from + different drivers. Instead, use the count for the driver_name. + type: Custom stabilityLevel: ALPHA + labels: + - driver_name - name: active_pods subsystem: kubelet help: The number of pods the kubelet considers active and which are being considered @@ -875,6 +911,15 @@ labels: - boundary - scope +- name: container_aligned_compute_resources_failure_count + subsystem: kubelet + help: Cumulative number of failures to allocate aligned compute resources to containers + by alignment type. + type: Counter + stabilityLevel: ALPHA + labels: + - boundary + - scope - name: kubelet_container_log_filesystem_used_bytes help: Bytes used by the container's logs on the filesystem. type: Custom @@ -884,6 +929,18 @@ - namespace - pod - container +- name: container_requested_resizes_total + subsystem: kubelet + help: Number of requested resizes, counted at the container level. Different resources + on the same container are counted separately. The 'requirement' label refers to + 'memory' or 'limits'; the 'operation' label can be one of 'add', 'remove', 'increase' + or 'decrease'. + type: Counter + stabilityLevel: ALPHA + labels: + - operation + - requirement + - resource - name: containers_per_pod_count subsystem: kubelet help: The number of containers per pod. @@ -895,6 +952,13 @@ - 4 - 8 - 16 +- name: cpu_manager_allocation_per_numa + subsystem: kubelet + help: Number of CPUs allocated per NUMA node + type: Gauge + stabilityLevel: ALPHA + labels: + - numa_node - name: cpu_manager_exclusive_cpu_allocation_count subsystem: kubelet help: The total number of CPUs exclusively allocated to containers running on this @@ -916,6 +980,14 @@ help: The size of the shared CPU pool for non-guaranteed QoS pods, in millicores. type: Gauge stabilityLevel: ALPHA +- name: cri_losing_support + subsystem: kubelet + help: the Kubernetes version that the currently running CRI implementation will + lose support on if not upgraded. + type: Gauge + stabilityLevel: ALPHA + labels: + - version - name: desired_pods subsystem: kubelet help: The number of pods the kubelet is being instructed to run. static is true @@ -1093,6 +1165,21 @@ - 1800 - 2700 - 3600 +- name: image_volume_mounted_errors_total + subsystem: kubelet + help: Number of failed image volume mounts. + type: Counter + stabilityLevel: ALPHA +- name: image_volume_mounted_succeed_total + subsystem: kubelet + help: Number of successful image volume mounts. + type: Counter + stabilityLevel: ALPHA +- name: image_volume_requested_total + subsystem: kubelet + help: Number of requested image volumes. + type: Counter + stabilityLevel: ALPHA - name: lifecycle_handler_http_fallbacks_total subsystem: kubelet help: The number of times lifecycle handlers successfully fell back to http from @@ -1216,6 +1303,54 @@ - 2.5 - 5 - 10 +- name: pod_deferred_accepted_resizes_total + subsystem: kubelet + help: Cumulative number of resizes that were accepted after being deferred. + type: Counter + stabilityLevel: ALPHA + labels: + - retry_trigger +- name: pod_in_progress_resizes + subsystem: kubelet + help: Number of in-progress resizes for pods. + type: Gauge + stabilityLevel: ALPHA +- name: pod_infeasible_resizes_total + subsystem: kubelet + help: Number of infeasible resizes for pods. + type: Counter + stabilityLevel: ALPHA + labels: + - reason_detail +- name: pod_pending_resizes + subsystem: kubelet + help: Number of pending resizes for pods. + type: Gauge + stabilityLevel: ALPHA + labels: + - reason +- name: pod_resize_duration_milliseconds + subsystem: kubelet + help: Duration in milliseconds to actuate a pod resize + type: Histogram + stabilityLevel: ALPHA + labels: + - success + buckets: + - 10 + - 50 + - 100 + - 500 + - 1000 + - 2000 + - 5000 + - 10000 + - 20000 + - 30000 + - 60000 + - 120000 + - 300000 + - 600000 - name: pod_resources_endpoint_errors_get subsystem: kubelet help: Number of requests to the PodResource Get endpoint which returned error. Broken @@ -1563,6 +1698,18 @@ help: Cumulative number of pods started type: Counter stabilityLevel: ALPHA +- name: started_user_namespaced_pods_errors_total + subsystem: kubelet + help: Cumulative number of errors when starting pods with user namespaces. This + metric will only be collected on Linux. + type: Counter + stabilityLevel: ALPHA +- name: started_user_namespaced_pods_total + subsystem: kubelet + help: Cumulative number of pods with user namespaces started. This metric will only + be collected on Linux. + type: Counter + stabilityLevel: ALPHA - name: topology_manager_admission_duration_ms subsystem: kubelet help: Duration in milliseconds to serve a pod admission request. @@ -1704,12 +1851,17 @@ - namespace - pod - probe_type +- name: scrape_error + help: 1 if there was an error while getting container metrics, 0 otherwise + type: Custom + deprecatedVersion: 1.29.0 + stabilityLevel: ALPHA - name: probe_total subsystem: prober help: Cumulative number of a liveness, readiness or startup probe for a container by result. type: Counter - stabilityLevel: ALPHA + stabilityLevel: BETA labels: - container - namespace @@ -1717,88 +1869,6 @@ - pod_uid - probe_type - result -- name: reconstruct_volume_operations_errors_total - help: The number of volumes that failed reconstruction from the operating system - during kubelet startup. - type: Counter - stabilityLevel: ALPHA -- name: reconstruct_volume_operations_total - help: The number of volumes that were attempted to be reconstructed from the operating - system during kubelet startup. This includes both successful and failed reconstruction. - type: Counter - stabilityLevel: ALPHA -- name: scrape_error - help: 1 if there was an error while getting container metrics, 0 otherwise - type: Custom - deprecatedVersion: 1.29.0 - stabilityLevel: ALPHA -- name: volume_manager_selinux_container_errors_total - help: Number of errors when kubelet cannot compute SELinux context for a container. - Kubelet can't start such a Pod then and it will retry, therefore value of this - metric may not represent the actual nr. of containers. - type: Gauge - stabilityLevel: ALPHA - labels: - - access_mode -- name: volume_manager_selinux_container_warnings_total - help: Number of errors when kubelet cannot compute SELinux context for a container - that are ignored. They will become real errors when SELinuxMountReadWriteOncePod - feature is expanded to all volume access modes. - type: Gauge - stabilityLevel: ALPHA - labels: - - access_mode -- name: volume_manager_selinux_pod_context_mismatch_errors_total - help: Number of errors when a Pod defines different SELinux contexts for its containers - that use the same volume. Kubelet can't start such a Pod then and it will retry, - therefore value of this metric may not represent the actual nr. of Pods. - type: Gauge - stabilityLevel: ALPHA - labels: - - access_mode -- name: volume_manager_selinux_pod_context_mismatch_warnings_total - help: Number of errors when a Pod defines different SELinux contexts for its containers - that use the same volume. They are not errors yet, but they will become real errors - when SELinuxMountReadWriteOncePod feature is expanded to all volume access modes. - type: Gauge - stabilityLevel: ALPHA - labels: - - access_mode -- name: volume_manager_selinux_volume_context_mismatch_errors_total - help: Number of errors when a Pod uses a volume that is already mounted with a different - SELinux context than the Pod needs. Kubelet can't start such a Pod then and it - will retry, therefore value of this metric may not represent the actual nr. of - Pods. - type: Gauge - stabilityLevel: ALPHA - labels: - - access_mode - - volume_plugin -- name: volume_manager_selinux_volume_context_mismatch_warnings_total - help: Number of errors when a Pod uses a volume that is already mounted with a different - SELinux context than the Pod needs. They are not errors yet, but they will become - real errors when SELinuxMountReadWriteOncePod feature is expanded to all volume - access modes. - type: Gauge - stabilityLevel: ALPHA - labels: - - access_mode - - volume_plugin -- name: volume_manager_selinux_volumes_admitted_total - help: Number of volumes whose SELinux context was fine and will be mounted with - mount -o context option. - type: Gauge - stabilityLevel: ALPHA - labels: - - access_mode - - volume_plugin -- name: volume_manager_total_volumes - help: Number of volumes in Volume Manager - type: Custom - stabilityLevel: ALPHA - labels: - - plugin_name - - state - name: container_cpu_usage_seconds_total help: Cumulative cpu time consumed by the container in core-seconds type: Custom @@ -1849,24 +1919,46 @@ help: 1 if there was an error while getting container metrics, 0 otherwise type: Custom stabilityLevel: STABLE -- name: csr_honored_duration_total - subsystem: certificates_registry - namespace: apiserver - help: Total number of issued CSRs with a requested duration that was honored, sliced - by signer (only kubernetes.io signer names are specifically identified) +- name: force_cleaned_failed_volume_operation_errors_total + help: The number of volumes that failed force cleanup after their reconstruction + failed during kubelet startup. + type: Counter + stabilityLevel: ALPHA +- name: force_cleaned_failed_volume_operations_total + help: The number of volumes that were force cleaned after their reconstruction failed + during kubelet startup. This includes both successful and failed cleanups. + type: Counter + stabilityLevel: ALPHA +- name: conntrack_reconciler_deleted_entries_total + subsystem: kubeproxy + help: Cumulative conntrack flows deleted by conntrack reconciler type: Counter stabilityLevel: ALPHA labels: - - signerName -- name: csr_requested_duration_total - subsystem: certificates_registry - namespace: apiserver - help: Total number of issued CSRs with a requested duration, sliced by signer (only - kubernetes.io signer names are specifically identified) - type: Counter + - ip_family +- name: conntrack_reconciler_sync_duration_seconds + subsystem: kubeproxy + help: ReconcileConntrackFlowsLatency latency in seconds + type: Histogram stabilityLevel: ALPHA labels: - - signerName + - ip_family + buckets: + - 0.001 + - 0.002 + - 0.004 + - 0.008 + - 0.016 + - 0.032 + - 0.064 + - 0.128 + - 0.256 + - 0.512 + - 1.024 + - 2.048 + - 4.096 + - 8.192 + - 16.384 - name: kubeproxy_iptables_ct_state_invalid_dropped_packets_total help: packets dropped by iptables to work around conntrack problems type: Custom @@ -1880,6 +1972,8 @@ help: In Cluster Network Programming Latency in seconds type: Histogram stabilityLevel: ALPHA + labels: + - ip_family buckets: - 0.25 - 0.5 @@ -1980,6 +2074,8 @@ help: SyncProxyRules latency in seconds for full resyncs type: Histogram stabilityLevel: ALPHA + labels: + - ip_family buckets: - 0.001 - 0.002 @@ -2001,6 +2097,8 @@ help: SyncProxyRules latency in seconds for partial resyncs type: Histogram stabilityLevel: ALPHA + labels: + - ip_family buckets: - 0.001 - 0.002 @@ -2022,6 +2120,8 @@ help: SyncProxyRules latency in seconds type: Histogram stabilityLevel: ALPHA + labels: + - ip_family buckets: - 0.001 - 0.002 @@ -2054,50 +2154,65 @@ type: Gauge stabilityLevel: ALPHA labels: + - ip_family - table - name: sync_proxy_rules_iptables_partial_restore_failures_total subsystem: kubeproxy help: Cumulative proxy iptables partial restore failures type: Counter stabilityLevel: ALPHA + labels: + - ip_family - name: sync_proxy_rules_iptables_restore_failures_total subsystem: kubeproxy help: Cumulative proxy iptables restore failures type: Counter stabilityLevel: ALPHA + labels: + - ip_family - name: sync_proxy_rules_iptables_total subsystem: kubeproxy help: Total number of iptables rules owned by kube-proxy type: Gauge stabilityLevel: ALPHA labels: + - ip_family - table - name: sync_proxy_rules_last_queued_timestamp_seconds subsystem: kubeproxy help: The last time a sync of proxy rules was queued type: Gauge stabilityLevel: ALPHA + labels: + - ip_family - name: sync_proxy_rules_last_timestamp_seconds subsystem: kubeproxy help: The last time proxy rules were successfully synced type: Gauge stabilityLevel: ALPHA + labels: + - ip_family - name: sync_proxy_rules_nftables_cleanup_failures_total subsystem: kubeproxy help: Cumulative proxy nftables cleanup failures type: Counter stabilityLevel: ALPHA + labels: + - ip_family - name: sync_proxy_rules_nftables_sync_failures_total subsystem: kubeproxy help: Cumulative proxy nftables sync failures type: Counter stabilityLevel: ALPHA + labels: + - ip_family - name: sync_proxy_rules_no_local_endpoints_total subsystem: kubeproxy help: Number of services with a Local traffic policy and no endpoints type: Gauge stabilityLevel: ALPHA labels: + - ip_family - traffic_policy - name: sync_proxy_rules_service_changes_pending subsystem: kubeproxy @@ -2109,6 +2224,101 @@ help: Cumulative proxy rules Service changes type: Counter stabilityLevel: ALPHA +- name: reconstruct_volume_operations_errors_total + help: The number of volumes that failed reconstruction from the operating system + during kubelet startup. + type: Counter + stabilityLevel: ALPHA +- name: reconstruct_volume_operations_total + help: The number of volumes that were attempted to be reconstructed from the operating + system during kubelet startup. This includes both successful and failed reconstruction. + type: Counter + stabilityLevel: ALPHA +- name: volume_manager_selinux_container_errors_total + help: Number of errors when kubelet cannot compute SELinux context for a container. + Kubelet can't start such a Pod then and it will retry, therefore value of this + metric may not represent the actual nr. of containers. + type: Gauge + stabilityLevel: ALPHA + labels: + - access_mode +- name: volume_manager_selinux_container_warnings_total + help: Number of errors when kubelet cannot compute SELinux context for a container + that are ignored. They will become real errors when SELinuxMountReadWriteOncePod + feature is expanded to all volume access modes. + type: Gauge + stabilityLevel: ALPHA + labels: + - access_mode +- name: volume_manager_selinux_pod_context_mismatch_errors_total + help: Number of errors when a Pod defines different SELinux contexts for its containers + that use the same volume. Kubelet can't start such a Pod then and it will retry, + therefore value of this metric may not represent the actual nr. of Pods. + type: Gauge + stabilityLevel: ALPHA + labels: + - access_mode +- name: volume_manager_selinux_pod_context_mismatch_warnings_total + help: Number of errors when a Pod defines different SELinux contexts for its containers + that use the same volume. They are not errors yet, but they will become real errors + when SELinuxMountReadWriteOncePod feature is expanded to all volume access modes. + type: Gauge + stabilityLevel: ALPHA + labels: + - access_mode +- name: volume_manager_selinux_volume_context_mismatch_errors_total + help: Number of errors when a Pod uses a volume that is already mounted with a different + SELinux context than the Pod needs. Kubelet can't start such a Pod then and it + will retry, therefore value of this metric may not represent the actual nr. of + Pods. + type: Gauge + stabilityLevel: ALPHA + labels: + - access_mode + - volume_plugin +- name: volume_manager_selinux_volume_context_mismatch_warnings_total + help: Number of errors when a Pod uses a volume that is already mounted with a different + SELinux context than the Pod needs. They are not errors yet, but they will become + real errors when SELinuxMountReadWriteOncePod feature is expanded to all volume + access modes. + type: Gauge + stabilityLevel: ALPHA + labels: + - access_mode + - volume_plugin +- name: volume_manager_selinux_volumes_admitted_total + help: Number of volumes whose SELinux context was fine and will be mounted with + mount -o context option. + type: Gauge + stabilityLevel: ALPHA + labels: + - access_mode + - volume_plugin +- name: volume_manager_total_volumes + help: Number of volumes in Volume Manager + type: Custom + stabilityLevel: ALPHA + labels: + - plugin_name + - state +- name: csr_honored_duration_total + subsystem: certificates_registry + namespace: apiserver + help: Total number of issued CSRs with a requested duration that was honored, sliced + by signer (only kubernetes.io signer names are specifically identified) + type: Counter + stabilityLevel: ALPHA + labels: + - signerName +- name: csr_requested_duration_total + subsystem: certificates_registry + namespace: apiserver + help: Total number of issued CSRs with a requested duration, sliced by signer (only + kubernetes.io signer names are specifically identified) + type: Counter + stabilityLevel: ALPHA + labels: + - signerName - name: ip_errors_total subsystem: clusterip_repair namespace: apiserver @@ -2308,6 +2518,45 @@ stabilityLevel: ALPHA labels: - code +- name: async_api_call_execution_duration_seconds + subsystem: scheduler + help: Duration in seconds for executing API call in the async dispatcher. + type: Histogram + stabilityLevel: ALPHA + labels: + - call_type + - result + buckets: + - 0.001 + - 0.002 + - 0.004 + - 0.008 + - 0.016 + - 0.032 + - 0.064 + - 0.128 + - 0.256 + - 0.512 + - 1.024 + - 2.048 + - 4.096 + - 8.192 + - 16.384 +- name: async_api_call_execution_total + subsystem: scheduler + help: Total number of API calls executed by the async dispatcher. + type: Counter + stabilityLevel: ALPHA + labels: + - call_type + - result +- name: cache_size + subsystem: scheduler + help: Number of nodes, pods, and assumed (bound) pods in the scheduler cache. + type: Gauge + stabilityLevel: ALPHA + labels: + - type - name: event_handling_duration_seconds subsystem: scheduler help: Event handling latency in seconds. @@ -2342,6 +2591,13 @@ stabilityLevel: ALPHA labels: - event +- name: pending_async_api_calls + subsystem: scheduler + help: Number of API calls currently pending in the async queue. + type: Gauge + stabilityLevel: ALPHA + labels: + - call_type - name: permit_wait_duration_seconds subsystem: scheduler help: Duration of waiting on permit. @@ -2470,13 +2726,6 @@ - 0.009852612533569338 - 0.014778918800354007 - 0.02216837820053101 -- name: scheduler_cache_size - subsystem: scheduler - help: Number of nodes, pods, and assumed (bound) pods in the scheduler cache. - type: Gauge - stabilityLevel: ALPHA - labels: - - type - name: scheduling_algorithm_duration_seconds subsystem: scheduler help: Scheduling algorithm latency in seconds @@ -2653,36 +2902,6 @@ - 4 - 8 - 16 -- name: pod_scheduling_duration_seconds - subsystem: scheduler - help: E2e latency for a pod being scheduled which may include multiple scheduling - attempts. - type: Histogram - deprecatedVersion: 1.29.0 - stabilityLevel: STABLE - labels: - - attempts - buckets: - - 0.01 - - 0.02 - - 0.04 - - 0.08 - - 0.16 - - 0.32 - - 0.64 - - 1.28 - - 2.56 - - 5.12 - - 10.24 - - 20.48 - - 40.96 - - 81.92 - - 163.84 - - 327.68 - - 655.36 - - 1310.72 - - 2621.44 - - 5242.88 - name: preemption_attempts_total subsystem: scheduler help: Total preemption attempts in the cluster till now @@ -2767,26 +2986,6 @@ - 120 - 300 - 600 -- name: graph_actions_duration_seconds - subsystem: node_authorizer - help: Histogram of duration of graph actions in node authorizer. - type: Histogram - stabilityLevel: ALPHA - labels: - - operation - buckets: - - 0.0001 - - 0.0002 - - 0.0004 - - 0.0008 - - 0.0016 - - 0.0032 - - 0.0064 - - 0.0128 - - 0.0256 - - 0.0512 - - 0.1024 - - 0.2048 - name: storage_operation_duration_seconds help: Storage operation duration type: Histogram @@ -2831,6 +3030,26 @@ - 120 - 300 - 600 +- name: graph_actions_duration_seconds + subsystem: node_authorizer + help: Histogram of duration of graph actions in node authorizer. + type: Histogram + stabilityLevel: ALPHA + labels: + - operation + buckets: + - 0.0001 + - 0.0002 + - 0.0004 + - 0.0008 + - 0.0016 + - 0.0032 + - 0.0064 + - 0.0128 + - 0.0256 + - 0.0512 + - 0.1024 + - 0.2048 - name: ratcheting_seconds subsystem: validation namespace: apiextensions_apiserver @@ -2849,24 +3068,6 @@ - 0.16384 - 0.65536 - 2.62144 -- name: apiextensions_openapi_v2_regeneration_count - help: Counter of OpenAPI v2 spec regeneration count broken down by causing CRD name - and reason. - type: Counter - stabilityLevel: ALPHA - labels: - - crd - - reason -- name: apiextensions_openapi_v3_regeneration_count - help: Counter of OpenAPI v3 spec regeneration count broken down by group, version, - causing CRD and reason. - type: Counter - stabilityLevel: ALPHA - labels: - - crd - - group - - reason - - version - name: conversion_webhook_duration_seconds namespace: apiserver help: Conversion webhook request latency @@ -2925,6 +3126,24 @@ - 4.096 - 8.192 - 16.384 +- name: apiextensions_openapi_v2_regeneration_count + help: Counter of OpenAPI v2 spec regeneration count broken down by causing CRD name + and reason. + type: Counter + stabilityLevel: ALPHA + labels: + - crd + - reason +- name: apiextensions_openapi_v3_regeneration_count + help: Counter of OpenAPI v3 spec regeneration count broken down by group, version, + causing CRD and reason. + type: Counter + stabilityLevel: ALPHA + labels: + - crd + - group + - reason + - version - name: match_condition_evaluation_errors_total subsystem: admission namespace: apiserver @@ -3028,6 +3247,34 @@ - operation - rejected - type +- name: check_duration_seconds + subsystem: mutating_admission_policy + namespace: apiserver + help: Mutation admission latency for individual mutation expressions in seconds, + labeled by policy and binding. + type: Histogram + stabilityLevel: ALPHA + labels: + - error_type + - policy + - policy_binding + buckets: + - 5e-07 + - 0.001 + - 0.01 + - 0.1 + - 1 +- name: check_total + subsystem: mutating_admission_policy + namespace: apiserver + help: Mutation admission policy check total, labeled by policy and further identified + by binding. + type: Counter + stabilityLevel: ALPHA + labels: + - error_type + - policy + - policy_binding - name: check_duration_seconds subsystem: validating_admission_policy namespace: apiserver @@ -3116,6 +3363,10 @@ - 2.5 - 10 - 25 +- name: aggregator_discovery_aggregation_count_total + help: Counter of number of times discovery was aggregated + type: Counter + stabilityLevel: ALPHA - name: error_total subsystem: apiserver_audit help: Counter of audit events that failed to be audited properly. Plugin identifies @@ -3212,6 +3463,14 @@ - 7.776e+06 - 1.5552e+07 - 3.1104e+07 +- name: current_inqueue_requests + subsystem: apiserver + help: Maximal number of queued requests in this apiserver per request kind in last + second. + type: Gauge + stabilityLevel: ALPHA + labels: + - request_kind - name: apiserver_delegated_authn_request_duration_seconds help: Request latency in seconds. Broken down by status code. type: Histogram @@ -3254,177 +3513,6 @@ stabilityLevel: ALPHA labels: - code -- name: active_fetch_count - subsystem: token_cache - namespace: authentication - type: Gauge - stabilityLevel: ALPHA - labels: - - status -- name: fetch_total - subsystem: token_cache - namespace: authentication - type: Counter - stabilityLevel: ALPHA - labels: - - status -- name: request_duration_seconds - subsystem: token_cache - namespace: authentication - type: Histogram - stabilityLevel: ALPHA - labels: - - status -- name: request_total - subsystem: token_cache - namespace: authentication - type: Counter - stabilityLevel: ALPHA - labels: - - status -- name: compilation_duration_seconds - subsystem: cel - namespace: apiserver - help: CEL compilation time in seconds. - type: Histogram - stabilityLevel: BETA -- name: evaluation_duration_seconds - subsystem: cel - namespace: apiserver - help: CEL evaluation time in seconds. - type: Histogram - stabilityLevel: BETA -- name: aggregator_discovery_aggregation_count_total - help: Counter of number of times discovery was aggregated - type: Counter - stabilityLevel: ALPHA -- name: automatic_reload_last_timestamp_seconds - subsystem: authentication_config_controller - namespace: apiserver - help: Timestamp of the last automatic reload of authentication configuration split - by status and apiserver identity. - type: Gauge - stabilityLevel: ALPHA - labels: - - apiserver_id_hash - - status -- name: automatic_reloads_total - subsystem: authentication_config_controller - namespace: apiserver - help: Total number of automatic reloads of authentication configuration split by - status and apiserver identity. - type: Counter - stabilityLevel: ALPHA - labels: - - apiserver_id_hash - - status -- name: automatic_reload_last_timestamp_seconds - subsystem: authorization_config_controller - namespace: apiserver - help: Timestamp of the last automatic reload of authorization configuration split - by status and apiserver identity. - type: Gauge - stabilityLevel: ALPHA - labels: - - apiserver_id_hash - - status -- name: automatic_reloads_total - subsystem: authorization_config_controller - namespace: apiserver - help: Total number of automatic reloads of authorization configuration split by - status and apiserver identity. - type: Counter - stabilityLevel: ALPHA - labels: - - apiserver_id_hash - - status -- name: current_inqueue_requests - subsystem: apiserver - help: Maximal number of queued requests in this apiserver per request kind in last - second. - type: Gauge - stabilityLevel: ALPHA - labels: - - request_kind -- name: dial_duration_seconds - subsystem: egress_dialer - namespace: apiserver - help: Dial latency histogram in seconds, labeled by the protocol (http-connect or - grpc), transport (tcp or uds) - type: Histogram - stabilityLevel: ALPHA - labels: - - protocol - - transport - buckets: - - 0.005 - - 0.025 - - 0.1 - - 0.5 - - 2.5 - - 12.5 -- name: dial_failure_count - subsystem: egress_dialer - namespace: apiserver - help: Dial failure count, labeled by the protocol (http-connect or grpc), transport - (tcp or uds), and stage (connect or proxy). The stage indicates at which stage - the dial failed - type: Counter - stabilityLevel: ALPHA - labels: - - protocol - - stage - - transport -- name: dial_start_total - subsystem: egress_dialer - namespace: apiserver - help: Dial starts, labeled by the protocol (http-connect or grpc) and transport - (tcp or uds). - type: Counter - stabilityLevel: ALPHA - labels: - - protocol - - transport -- name: automatic_reload_failures_total - subsystem: encryption_config_controller - namespace: apiserver - help: Total number of failed automatic reloads of encryption configuration split - by apiserver identity. - type: Counter - deprecatedVersion: 1.30.0 - stabilityLevel: ALPHA - labels: - - apiserver_id_hash -- name: automatic_reload_last_timestamp_seconds - subsystem: encryption_config_controller - namespace: apiserver - help: Timestamp of the last successful or failed automatic reload of encryption - configuration split by apiserver identity. - type: Gauge - stabilityLevel: ALPHA - labels: - - apiserver_id_hash - - status -- name: automatic_reload_success_total - subsystem: encryption_config_controller - namespace: apiserver - help: Total number of successful automatic reloads of encryption configuration split - by apiserver identity. - type: Counter - deprecatedVersion: 1.30.0 - stabilityLevel: ALPHA - labels: - - apiserver_id_hash -- name: automatic_reloads_total - subsystem: encryption_config_controller - namespace: apiserver - help: Total number of reload successes and failures of encryption configuration - split by apiserver identity. - type: Counter - stabilityLevel: ALPHA - labels: - - apiserver_id_hash - - status - name: request_aborts_total subsystem: apiserver help: Number of requests which apiserver aborted possibly due to a timeout, for @@ -3444,6 +3532,7 @@ type: Histogram stabilityLevel: ALPHA labels: + - group - resource - verb buckets: @@ -3622,6 +3711,7 @@ type: Counter stabilityLevel: ALPHA labels: + - group - resource - subresource - verb @@ -3637,7 +3727,7 @@ stabilityLevel: ALPHA labels: - group - - kind + - resource - version buckets: - 1024 @@ -3655,7 +3745,7 @@ stabilityLevel: ALPHA labels: - group - - kind + - resource - version - name: watch_list_duration_seconds subsystem: apiserver @@ -3720,6 +3810,34 @@ - 4.096 - 8.192 - 16.384 +- name: active_fetch_count + subsystem: token_cache + namespace: authentication + type: Gauge + stabilityLevel: ALPHA + labels: + - status +- name: fetch_total + subsystem: token_cache + namespace: authentication + type: Counter + stabilityLevel: ALPHA + labels: + - status +- name: request_duration_seconds + subsystem: token_cache + namespace: authentication + type: Histogram + stabilityLevel: ALPHA + labels: + - status +- name: request_total + subsystem: token_cache + namespace: authentication + type: Counter + stabilityLevel: ALPHA + labels: + - status - name: authorization_attempts_total help: Counter of authorization attempts broken down by result. It can be either 'allowed', 'denied', 'no-opinion' or 'error'. @@ -3777,6 +3895,18 @@ - 30 - 45 - 60 +- name: compilation_duration_seconds + subsystem: cel + namespace: apiserver + help: CEL compilation time in seconds. + type: Histogram + stabilityLevel: BETA +- name: evaluation_duration_seconds + subsystem: cel + namespace: apiserver + help: CEL evaluation time in seconds. + type: Histogram + stabilityLevel: BETA - name: current_inflight_requests subsystem: apiserver help: Maximal number of currently used inflight request limit of this apiserver @@ -3888,92 +4018,115 @@ - 1e+07 - 1e+08 - 1e+09 -- name: jwt_authenticator_latency_seconds - subsystem: authentication - namespace: apiserver - help: Latency of jwt authentication operations in seconds. This is the time spent - authenticating a token for cache miss only (i.e. when the token is not found in - the cache). - type: Histogram +- name: apiserver_authentication_config_controller_last_config_info + help: Information about the last applied authentication configuration with hash + as label, split by apiserver identity. + type: Custom stabilityLevel: ALPHA labels: - - jwt_issuer_hash - - result - buckets: - - 0.001 - - 0.005 - - 0.01 - - 0.025 - - 0.05 - - 0.1 - - 0.25 - - 0.5 - - 1 - - 2.5 - - 5 - - 10 -- name: webhook_duration_seconds - subsystem: authorization - namespace: apiserver - help: Request latency in seconds. - type: Histogram + - apiserver_id_hash + - hash +- name: apiserver_authorization_config_controller_last_config_info + help: Information about the last applied authorization configuration with hash as + label, split by apiserver identity. + type: Custom stabilityLevel: ALPHA labels: - - name - - result - buckets: - - 0.005 - - 0.01 - - 0.025 - - 0.05 - - 0.1 - - 0.25 - - 0.5 - - 1 - - 2.5 - - 5 - - 10 -- name: webhook_evaluations_fail_open_total - subsystem: authorization - namespace: apiserver - help: NoOpinion results due to webhook timeout or error. - type: Counter - stabilityLevel: ALPHA - labels: - - name - - result -- name: webhook_evaluations_total - subsystem: authorization - namespace: apiserver - help: Round-trips to authorization webhooks. - type: Counter - stabilityLevel: ALPHA - labels: - - name - - result + - apiserver_id_hash + - hash - name: cache_list_fetched_objects_total namespace: apiserver help: Number of objects read from watch cache in the course of serving a LIST request type: Counter stabilityLevel: ALPHA labels: + - group - index - - resource_prefix + - resource - name: cache_list_returned_objects_total namespace: apiserver help: Number of objects returned for a LIST request from watch cache type: Counter stabilityLevel: ALPHA labels: - - resource_prefix + - group + - resource - name: cache_list_total namespace: apiserver help: Number of LIST requests served from watch cache type: Counter stabilityLevel: ALPHA labels: + - group - index - - resource_prefix + - resource +- name: dial_duration_seconds + subsystem: egress_dialer + namespace: apiserver + help: Dial latency histogram in seconds, labeled by the protocol (http-connect or + grpc), transport (tcp or uds) + type: Histogram + stabilityLevel: ALPHA + labels: + - protocol + - transport + buckets: + - 0.005 + - 0.025 + - 0.1 + - 0.5 + - 2.5 + - 12.5 +- name: dial_failure_count + subsystem: egress_dialer + namespace: apiserver + help: Dial failure count, labeled by the protocol (http-connect or grpc), transport + (tcp or uds), and stage (connect or proxy). The stage indicates at which stage + the dial failed + type: Counter + stabilityLevel: ALPHA + labels: + - protocol + - stage + - transport +- name: dial_start_total + subsystem: egress_dialer + namespace: apiserver + help: Dial starts, labeled by the protocol (http-connect or grpc) and transport + (tcp or uds). + type: Counter + stabilityLevel: ALPHA + labels: + - protocol + - transport +- name: automatic_reload_last_timestamp_seconds + subsystem: encryption_config_controller + namespace: apiserver + help: Timestamp of the last successful or failed automatic reload of encryption + configuration split by apiserver identity. + type: Gauge + stabilityLevel: ALPHA + labels: + - apiserver_id_hash + - status +- name: automatic_reloads_total + subsystem: encryption_config_controller + namespace: apiserver + help: Total number of reload successes and failures of encryption configuration + split by apiserver identity. + type: Counter + stabilityLevel: ALPHA + labels: + - apiserver_id_hash + - status +- name: apiserver_encryption_config_controller_last_config_info + help: Information about the last applied encryption configuration with hash as label, + split by apiserver identity. + type: Custom + stabilityLevel: ALPHA + labels: + - apiserver_id_hash + - hash - name: dek_cache_fill_percent subsystem: envelope_encryption namespace: apiserver @@ -4083,6 +4236,464 @@ - 13.1072 - 26.2144 - 52.4288 +- name: init_events_total + namespace: apiserver + help: Counter of init events processed in watch cache broken by resource type. + type: Counter + stabilityLevel: ALPHA + labels: + - group + - resource +- name: apiserver_resource_objects + help: Number of stored objects at the time of last check split by kind. In case + of a fetching error, the value will be -1. + type: Gauge + stabilityLevel: ALPHA + labels: + - group + - resource +- name: apiserver_resource_size_estimate_bytes + help: Estimated size of stored objects in database. Estimate is based on sum of + last observed sizes of serialized objects. In case of a fetching error, the value + will be -1. + type: Gauge + stabilityLevel: ALPHA + labels: + - group + - resource +- name: storage_consistency_checks_total + namespace: apiserver + help: Counter for status of consistency checks between etcd and watch cache + type: Counter + stabilityLevel: ALPHA + labels: + - group + - resource + - status +- name: data_key_generation_duration_seconds + subsystem: storage + namespace: apiserver + help: Latencies in seconds of data encryption key(DEK) generation operations. + type: Histogram + stabilityLevel: ALPHA + buckets: + - 5e-06 + - 1e-05 + - 2e-05 + - 4e-05 + - 8e-05 + - 0.00016 + - 0.00032 + - 0.00064 + - 0.00128 + - 0.00256 + - 0.00512 + - 0.01024 + - 0.02048 + - 0.04096 +- name: data_key_generation_failures_total + subsystem: storage + namespace: apiserver + help: Total number of failed data encryption key(DEK) generation operations. + type: Counter + stabilityLevel: ALPHA +- name: storage_db_total_size_in_bytes + subsystem: apiserver + help: Total size of the storage database file physically allocated in bytes. + type: Gauge + deprecatedVersion: 1.28.0 + stabilityLevel: ALPHA + labels: + - endpoint +- name: storage_decode_errors_total + namespace: apiserver + help: Number of stored object decode errors split by object type + type: Counter + stabilityLevel: ALPHA + labels: + - group + - resource +- name: envelope_transformation_cache_misses_total + subsystem: storage + namespace: apiserver + help: Total number of cache misses while accessing key decryption key(KEK). + type: Counter + stabilityLevel: ALPHA +- name: storage_events_received_total + subsystem: apiserver + help: Number of etcd events received split by kind. + type: Counter + stabilityLevel: ALPHA + labels: + - group + - resource +- name: apiserver_storage_list_evaluated_objects_total + help: Number of objects tested in the course of serving a LIST request from storage + type: Counter + stabilityLevel: ALPHA + labels: + - group + - resource +- name: apiserver_storage_list_fetched_objects_total + help: Number of objects read from storage in the course of serving a LIST request + type: Counter + stabilityLevel: ALPHA + labels: + - group + - resource +- name: apiserver_storage_list_returned_objects_total + help: Number of objects returned for a LIST request from storage + type: Counter + stabilityLevel: ALPHA + labels: + - group + - resource +- name: apiserver_storage_list_total + help: Number of LIST requests served from storage + type: Counter + stabilityLevel: ALPHA + labels: + - group + - resource +- name: transformation_duration_seconds + subsystem: storage + namespace: apiserver + help: Latencies in seconds of value transformation operations. + type: Histogram + stabilityLevel: ALPHA + labels: + - transformation_type + - transformer_prefix + buckets: + - 5e-06 + - 1e-05 + - 2e-05 + - 4e-05 + - 8e-05 + - 0.00016 + - 0.00032 + - 0.00064 + - 0.00128 + - 0.00256 + - 0.00512 + - 0.01024 + - 0.02048 + - 0.04096 + - 0.08192 + - 0.16384 + - 0.32768 + - 0.65536 + - 1.31072 + - 2.62144 + - 5.24288 + - 10.48576 + - 20.97152 + - 41.94304 + - 83.88608 +- name: transformation_operations_total + subsystem: storage + namespace: apiserver + help: Total number of transformations. Successful transformation will have a status + 'OK' and a varied status string when the transformation fails. The status, resource, + and transformation_type fields can be used for alerting purposes. For example, + you can monitor for encryption/decryption failures using the transformation_type + (e.g., from_storage for decryption and to_storage for encryption). Additionally, + these fields can be used to ensure that the correct transformers are applied to + each resource. + type: Counter + stabilityLevel: ALPHA + labels: + - resource + - status + - transformation_type + - transformer_prefix +- name: terminated_watchers_total + namespace: apiserver + help: Counter of watchers closed due to unresponsiveness broken by resource type. + type: Counter + stabilityLevel: ALPHA + labels: + - group + - resource +- name: consistent_read_total + subsystem: watch_cache + namespace: apiserver + help: Counter for consistent reads from cache. + type: Counter + stabilityLevel: ALPHA + labels: + - fallback + - group + - resource + - success +- name: events_dispatched_total + subsystem: watch_cache + namespace: apiserver + help: Counter of events dispatched in watch cache broken by resource type. + type: Counter + stabilityLevel: ALPHA + labels: + - group + - resource +- name: events_received_total + subsystem: watch_cache + namespace: apiserver + help: Counter of events received in watch cache broken by resource type. + type: Counter + stabilityLevel: ALPHA + labels: + - group + - resource +- name: initializations_total + subsystem: watch_cache + namespace: apiserver + help: Counter of watch cache initializations broken by resource type. + type: Counter + stabilityLevel: ALPHA + labels: + - group + - resource +- name: read_wait_seconds + subsystem: watch_cache + namespace: apiserver + help: Histogram of time spent waiting for a watch cache to become fresh. + type: Histogram + stabilityLevel: ALPHA + labels: + - group + - resource + buckets: + - 0.005 + - 0.025 + - 0.05 + - 0.1 + - 0.2 + - 0.4 + - 0.6 + - 0.8 + - 1 + - 1.25 + - 1.5 + - 2 + - 3 +- name: resource_version + subsystem: watch_cache + namespace: apiserver + help: Current resource version of watch cache broken by resource type. + type: Gauge + stabilityLevel: ALPHA + labels: + - group + - resource +- name: etcd_bookmark_counts + help: Number of etcd bookmarks (progress notify events) split by kind. + type: Gauge + stabilityLevel: ALPHA + labels: + - group + - resource +- name: etcd_lease_object_counts + help: Number of objects attached to a single etcd lease. + type: Histogram + stabilityLevel: ALPHA + buckets: + - 10 + - 50 + - 100 + - 500 + - 1000 + - 2500 + - 5000 +- name: etcd_request_duration_seconds + help: Etcd request latency in seconds for each operation and object type. + type: Histogram + stabilityLevel: ALPHA + labels: + - group + - operation + - resource + buckets: + - 0.005 + - 0.025 + - 0.05 + - 0.1 + - 0.2 + - 0.4 + - 0.6 + - 0.8 + - 1 + - 1.25 + - 1.5 + - 2 + - 3 + - 4 + - 5 + - 6 + - 8 + - 10 + - 15 + - 20 + - 30 + - 45 + - 60 +- name: etcd_request_errors_total + help: Etcd failed request counts for each operation and object type. + type: Counter + stabilityLevel: ALPHA + labels: + - group + - operation + - resource +- name: etcd_requests_total + help: Etcd request counts for each operation and object type. + type: Counter + stabilityLevel: ALPHA + labels: + - group + - operation + - resource +- name: capacity + subsystem: watch_cache + help: Total capacity of watch cache broken by resource type. + type: Gauge + stabilityLevel: ALPHA + labels: + - group + - resource +- name: capacity_decrease_total + subsystem: watch_cache + help: Total number of watch cache capacity decrease events broken by resource type. + type: Counter + stabilityLevel: ALPHA + labels: + - group + - resource +- name: capacity_increase_total + subsystem: watch_cache + help: Total number of watch cache capacity increase events broken by resource type. + type: Counter + stabilityLevel: ALPHA + labels: + - group + - resource +- name: automatic_reload_last_timestamp_seconds + subsystem: authentication_config_controller + namespace: apiserver + help: Timestamp of the last automatic reload of authentication configuration split + by status and apiserver identity. + type: Gauge + stabilityLevel: BETA + labels: + - apiserver_id_hash + - status +- name: automatic_reloads_total + subsystem: authentication_config_controller + namespace: apiserver + help: Total number of automatic reloads of authentication configuration split by + status and apiserver identity. + type: Counter + stabilityLevel: BETA + labels: + - apiserver_id_hash + - status +- name: automatic_reload_last_timestamp_seconds + subsystem: authorization_config_controller + namespace: apiserver + help: Timestamp of the last automatic reload of authorization configuration split + by status and apiserver identity. + type: Gauge + stabilityLevel: BETA + labels: + - apiserver_id_hash + - status +- name: automatic_reloads_total + subsystem: authorization_config_controller + namespace: apiserver + help: Total number of automatic reloads of authorization configuration split by + status and apiserver identity. + type: Counter + stabilityLevel: BETA + labels: + - apiserver_id_hash + - status +- name: apiserver_storage_objects + help: '[DEPRECATED, consider using apiserver_resource_objects instead] Number of + stored objects at the time of last check split by kind. In case of a fetching + error, the value will be -1.' + type: Gauge + deprecatedVersion: 1.34.0 + stabilityLevel: STABLE + labels: + - resource +- name: apiserver_storage_size_bytes + help: Size of the storage database file physically allocated in bytes. + type: Custom + stabilityLevel: STABLE + labels: + - storage_cluster_id +- name: jwt_authenticator_latency_seconds + subsystem: authentication + namespace: apiserver + help: Latency of jwt authentication operations in seconds. This is the time spent + authenticating a token for cache miss only (i.e. when the token is not found in + the cache). + type: Histogram + stabilityLevel: ALPHA + labels: + - jwt_issuer_hash + - result + buckets: + - 0.001 + - 0.005 + - 0.01 + - 0.025 + - 0.05 + - 0.1 + - 0.25 + - 0.5 + - 1 + - 2.5 + - 5 + - 10 +- name: webhook_duration_seconds + subsystem: authorization + namespace: apiserver + help: Request latency in seconds. + type: Histogram + stabilityLevel: ALPHA + labels: + - name + - result + buckets: + - 0.005 + - 0.01 + - 0.025 + - 0.05 + - 0.1 + - 0.25 + - 0.5 + - 1 + - 2.5 + - 5 + - 10 +- name: webhook_evaluations_fail_open_total + subsystem: authorization + namespace: apiserver + help: NoOpinion results due to webhook timeout or error. + type: Counter + stabilityLevel: ALPHA + labels: + - name + - result +- name: webhook_evaluations_total + subsystem: authorization + namespace: apiserver + help: Round-trips to authorization webhooks. + type: Counter + stabilityLevel: ALPHA + labels: + - name + - result - name: current_inqueue_seats subsystem: flowcontrol namespace: apiserver @@ -4418,14 +5029,11 @@ - 1 - 2 - 4 - - 10 -- name: init_events_total - namespace: apiserver - help: Counter of init events processed in watch cache broken by resource type. - type: Counter - stabilityLevel: ALPHA - labels: - - resource + - 8 + - 16 + - 32 + - 64 + - 100 - name: rerouted_request_total subsystem: apiserver help: Total number of requests that were proxied to a peer kube apiserver because @@ -4434,137 +5042,6 @@ stabilityLevel: ALPHA labels: - code -- name: data_key_generation_duration_seconds - subsystem: storage - namespace: apiserver - help: Latencies in seconds of data encryption key(DEK) generation operations. - type: Histogram - stabilityLevel: ALPHA - buckets: - - 5e-06 - - 1e-05 - - 2e-05 - - 4e-05 - - 8e-05 - - 0.00016 - - 0.00032 - - 0.00064 - - 0.00128 - - 0.00256 - - 0.00512 - - 0.01024 - - 0.02048 - - 0.04096 -- name: data_key_generation_failures_total - subsystem: storage - namespace: apiserver - help: Total number of failed data encryption key(DEK) generation operations. - type: Counter - stabilityLevel: ALPHA -- name: storage_db_total_size_in_bytes - subsystem: apiserver - help: Total size of the storage database file physically allocated in bytes. - type: Gauge - deprecatedVersion: 1.28.0 - stabilityLevel: ALPHA - labels: - - endpoint -- name: storage_decode_errors_total - namespace: apiserver - help: Number of stored object decode errors split by object type - type: Counter - stabilityLevel: ALPHA - labels: - - resource -- name: envelope_transformation_cache_misses_total - subsystem: storage - namespace: apiserver - help: Total number of cache misses while accessing key decryption key(KEK). - type: Counter - stabilityLevel: ALPHA -- name: storage_events_received_total - subsystem: apiserver - help: Number of etcd events received split by kind. - type: Counter - stabilityLevel: ALPHA - labels: - - resource -- name: apiserver_storage_list_evaluated_objects_total - help: Number of objects tested in the course of serving a LIST request from storage - type: Counter - stabilityLevel: ALPHA - labels: - - resource -- name: apiserver_storage_list_fetched_objects_total - help: Number of objects read from storage in the course of serving a LIST request - type: Counter - stabilityLevel: ALPHA - labels: - - resource -- name: apiserver_storage_list_returned_objects_total - help: Number of objects returned for a LIST request from storage - type: Counter - stabilityLevel: ALPHA - labels: - - resource -- name: apiserver_storage_list_total - help: Number of LIST requests served from storage - type: Counter - stabilityLevel: ALPHA - labels: - - resource -- name: transformation_duration_seconds - subsystem: storage - namespace: apiserver - help: Latencies in seconds of value transformation operations. - type: Histogram - stabilityLevel: ALPHA - labels: - - transformation_type - - transformer_prefix - buckets: - - 5e-06 - - 1e-05 - - 2e-05 - - 4e-05 - - 8e-05 - - 0.00016 - - 0.00032 - - 0.00064 - - 0.00128 - - 0.00256 - - 0.00512 - - 0.01024 - - 0.02048 - - 0.04096 - - 0.08192 - - 0.16384 - - 0.32768 - - 0.65536 - - 1.31072 - - 2.62144 - - 5.24288 - - 10.48576 - - 20.97152 - - 41.94304 - - 83.88608 -- name: transformation_operations_total - subsystem: storage - namespace: apiserver - help: Total number of transformations. Successful transformation will have a status - 'OK' and a varied status string when the transformation fails. The status, resource, - and transformation_type fields can be used for alerting purposes. For example, - you can monitor for encryption/decryption failures using the transformation_type - (e.g., from_storage for decryption and to_storage for encryption). Additionally, - these fields can be used to ensure that the correct transformers are applied to - each resource. - type: Counter - stabilityLevel: ALPHA - labels: - - resource - - status - - transformation_type - - transformer_prefix - name: stream_translator_requests_total subsystem: apiserver help: Total number of requests that were handled by the StreamTranslatorProxy, which @@ -4581,77 +5058,6 @@ stabilityLevel: ALPHA labels: - code -- name: terminated_watchers_total - namespace: apiserver - help: Counter of watchers closed due to unresponsiveness broken by resource type. - type: Counter - stabilityLevel: ALPHA - labels: - - resource -- name: consistent_read_total - subsystem: watch_cache - namespace: apiserver - help: Counter for consistent reads from cache. - type: Counter - stabilityLevel: ALPHA - labels: - - fallback - - resource - - success -- name: events_dispatched_total - subsystem: watch_cache - namespace: apiserver - help: Counter of events dispatched in watch cache broken by resource type. - type: Counter - stabilityLevel: ALPHA - labels: - - resource -- name: events_received_total - subsystem: watch_cache - namespace: apiserver - help: Counter of events received in watch cache broken by resource type. - type: Counter - stabilityLevel: ALPHA - labels: - - resource -- name: initializations_total - subsystem: watch_cache - namespace: apiserver - help: Counter of watch cache initializations broken by resource type. - type: Counter - stabilityLevel: ALPHA - labels: - - resource -- name: read_wait_seconds - subsystem: watch_cache - namespace: apiserver - help: Histogram of time spent waiting for a watch cache to become fresh. - type: Histogram - stabilityLevel: ALPHA - labels: - - resource - buckets: - - 0.005 - - 0.025 - - 0.05 - - 0.1 - - 0.2 - - 0.4 - - 0.6 - - 0.8 - - 1 - - 1.25 - - 1.5 - - 2 - - 3 -- name: resource_version - subsystem: watch_cache - namespace: apiserver - help: Current resource version of watch cache broken by resource type. - type: Gauge - stabilityLevel: ALPHA - labels: - - resource - name: x509_insecure_sha1_total subsystem: webhooks namespace: apiserver @@ -4668,90 +5074,6 @@ SAN extension missing (either/or, based on the runtime environment) type: Counter stabilityLevel: ALPHA -- name: etcd_bookmark_counts - help: Number of etcd bookmarks (progress notify events) split by kind. - type: Gauge - stabilityLevel: ALPHA - labels: - - resource -- name: etcd_lease_object_counts - help: Number of objects attached to a single etcd lease. - type: Histogram - stabilityLevel: ALPHA - buckets: - - 10 - - 50 - - 100 - - 500 - - 1000 - - 2500 - - 5000 -- name: etcd_request_duration_seconds - help: Etcd request latency in seconds for each operation and object type. - type: Histogram - stabilityLevel: ALPHA - labels: - - operation - - type - buckets: - - 0.005 - - 0.025 - - 0.05 - - 0.1 - - 0.2 - - 0.4 - - 0.6 - - 0.8 - - 1 - - 1.25 - - 1.5 - - 2 - - 3 - - 4 - - 5 - - 6 - - 8 - - 10 - - 15 - - 20 - - 30 - - 45 - - 60 -- name: etcd_request_errors_total - help: Etcd failed request counts for each operation and object type. - type: Counter - stabilityLevel: ALPHA - labels: - - operation - - type -- name: etcd_requests_total - help: Etcd request counts for each operation and object type. - type: Counter - stabilityLevel: ALPHA - labels: - - operation - - type -- name: capacity - subsystem: watch_cache - help: Total capacity of watch cache broken by resource type. - type: Gauge - stabilityLevel: ALPHA - labels: - - resource -- name: capacity_decrease_total - subsystem: watch_cache - help: Total number of watch cache capacity decrease events broken by resource type. - type: Counter - stabilityLevel: ALPHA - labels: - - resource -- name: capacity_increase_total - subsystem: watch_cache - help: Total number of watch cache capacity increase events broken by resource type. - type: Counter - stabilityLevel: ALPHA - labels: - - resource - name: current_executing_requests subsystem: flowcontrol namespace: apiserver @@ -4834,19 +5156,19 @@ - 10 - 15 - 30 -- name: apiserver_storage_objects - help: Number of stored objects at the time of last check split by kind. In case - of a fetching error, the value will be -1. - type: Gauge - stabilityLevel: STABLE - labels: - - resource -- name: apiserver_storage_size_bytes - help: Size of the storage database file physically allocated in bytes. - type: Custom - stabilityLevel: STABLE - labels: - - storage_cluster_id +- name: declarative_validation_mismatch_total + subsystem: validation + namespace: apiserver + help: Number of times declarative validation results differed from handwritten validation + results for core types. + type: Counter + stabilityLevel: BETA +- name: declarative_validation_panic_total + subsystem: validation + namespace: apiserver + help: Number of times declarative validation has panicked during validation. + type: Counter + stabilityLevel: BETA - name: request_duration_seconds subsystem: cloud_provider_webhook help: Request latency in seconds. Broken down by status code. @@ -4953,92 +5275,6 @@ - 4096 - 8192 - 16384 -- name: changes - subsystem: endpoint_slice_controller - help: Number of EndpointSlice changes - type: Counter - stabilityLevel: ALPHA - labels: - - operation -- name: desired_endpoint_slices - subsystem: endpoint_slice_controller - help: Number of EndpointSlices that would exist with perfect endpoint allocation - type: Gauge - stabilityLevel: ALPHA -- name: endpoints_added_per_sync - subsystem: endpoint_slice_controller - help: Number of endpoints added on each Service sync - type: Histogram - stabilityLevel: ALPHA - buckets: - - 2 - - 4 - - 8 - - 16 - - 32 - - 64 - - 128 - - 256 - - 512 - - 1024 - - 2048 - - 4096 - - 8192 - - 16384 - - 32768 -- name: endpoints_desired - subsystem: endpoint_slice_controller - help: Number of endpoints desired - type: Gauge - stabilityLevel: ALPHA -- name: endpoints_removed_per_sync - subsystem: endpoint_slice_controller - help: Number of endpoints removed on each Service sync - type: Histogram - stabilityLevel: ALPHA - buckets: - - 2 - - 4 - - 8 - - 16 - - 32 - - 64 - - 128 - - 256 - - 512 - - 1024 - - 2048 - - 4096 - - 8192 - - 16384 - - 32768 -- name: endpointslices_changed_per_sync - subsystem: endpoint_slice_controller - help: Number of EndpointSlices changed on each Service sync - type: Histogram - stabilityLevel: ALPHA - labels: - - topology - - traffic_distribution -- name: num_endpoint_slices - subsystem: endpoint_slice_controller - help: Number of EndpointSlices - type: Gauge - stabilityLevel: ALPHA -- name: services_count_by_traffic_distribution - subsystem: endpoint_slice_controller - help: Number of Services using some specific trafficDistribution - type: Gauge - stabilityLevel: ALPHA - labels: - - traffic_distribution -- name: syncs - subsystem: endpoint_slice_controller - help: Number of EndpointSlice syncs - type: Counter - stabilityLevel: ALPHA - labels: - - result - name: kubernetes_build_info help: A metric with a constant '1' value labeled by major, minor, git version, git commit, git tree state, build date, Go version, and compiler from which Kubernetes @@ -5236,6 +5472,16 @@ labels: - manager - name +- name: version_info + help: Provides the compatibility version info of the component. The component label + is the name of the component, usually kube, but is relevant for aggregated-apiservers. + type: Gauge + stabilityLevel: ALPHA + labels: + - binary + - component + - emulation + - min_compat - name: adds_total subsystem: workqueue help: Total number of adds handled by workqueue @@ -5364,21 +5610,6 @@ stabilityLevel: ALPHA labels: - reason -- name: aggregator_unavailable_apiservice - help: Gauge of APIServices which are marked as unavailable broken down by APIService - name. - type: Custom - stabilityLevel: ALPHA - labels: - - name -- name: aggregator_unavailable_apiservice_total - help: Counter of APIServices which are marked as unavailable broken down by APIService - name and reason. - type: Counter - stabilityLevel: ALPHA - labels: - - name - - reason - name: x509_insecure_sha1_total subsystem: kube_aggregator namespace: apiserver @@ -5395,6 +5626,107 @@ SAN extension missing (either/or, based on the runtime environment) type: Counter stabilityLevel: ALPHA +- name: changes + subsystem: endpoint_slice_controller + help: Number of EndpointSlice changes + type: Counter + stabilityLevel: ALPHA + labels: + - operation +- name: desired_endpoint_slices + subsystem: endpoint_slice_controller + help: Number of EndpointSlices that would exist with perfect endpoint allocation + type: Gauge + stabilityLevel: ALPHA +- name: endpoints_added_per_sync + subsystem: endpoint_slice_controller + help: Number of endpoints added on each Service sync + type: Histogram + stabilityLevel: ALPHA + buckets: + - 2 + - 4 + - 8 + - 16 + - 32 + - 64 + - 128 + - 256 + - 512 + - 1024 + - 2048 + - 4096 + - 8192 + - 16384 + - 32768 +- name: endpoints_desired + subsystem: endpoint_slice_controller + help: Number of endpoints desired + type: Gauge + stabilityLevel: ALPHA +- name: endpoints_removed_per_sync + subsystem: endpoint_slice_controller + help: Number of endpoints removed on each Service sync + type: Histogram + stabilityLevel: ALPHA + buckets: + - 2 + - 4 + - 8 + - 16 + - 32 + - 64 + - 128 + - 256 + - 512 + - 1024 + - 2048 + - 4096 + - 8192 + - 16384 + - 32768 +- name: endpointslices_changed_per_sync + subsystem: endpoint_slice_controller + help: Number of EndpointSlices changed on each Service sync + type: Histogram + stabilityLevel: ALPHA + labels: + - topology + - traffic_distribution +- name: num_endpoint_slices + subsystem: endpoint_slice_controller + help: Number of EndpointSlices + type: Gauge + stabilityLevel: ALPHA +- name: services_count_by_traffic_distribution + subsystem: endpoint_slice_controller + help: Number of Services using some specific trafficDistribution + type: Gauge + stabilityLevel: ALPHA + labels: + - traffic_distribution +- name: syncs + subsystem: endpoint_slice_controller + help: Number of EndpointSlice syncs + type: Counter + stabilityLevel: ALPHA + labels: + - result +- name: aggregator_unavailable_apiservice + help: Gauge of APIServices which are marked as unavailable broken down by APIService + name. + type: Custom + stabilityLevel: ALPHA + labels: + - name +- name: aggregator_unavailable_apiservice_total + help: Counter of APIServices which are marked as unavailable broken down by APIService + name and reason. + type: Counter + stabilityLevel: ALPHA + labels: + - name + - reason - name: pod_security_errors_total help: Number of errors preventing normal evaluation. Non-fatal errors may result in the latest restricted profile being used for evaluation. diff --git a/test/instrumentation/documentation/documentation.md b/test/instrumentation/documentation/documentation.md index bb64782fa53..017b4a748a2 100644 --- a/test/instrumentation/documentation/documentation.md +++ b/test/instrumentation/documentation/documentation.md @@ -6,10 +6,10 @@ description: >- Details of the metric data that Kubernetes components export. --- -## Metrics (v1.32) +## Metrics (v1.34) - - + + This page details the metrics that different Kubernetes components export. You can query the metrics endpoint for these components using an HTTP scrape, and fetch the current metrics data in Prometheus format. @@ -82,11 +82,11 @@ Stable metrics observe strict API contracts and no labels can be added or remove
  • componentgroupresourcescopesubresourceverbversion
    • apiserver_storage_objects
    -
    Number of stored objects at the time of last check split by kind. In case of a fetching error, the value will be -1.
    +
    [DEPRECATED, consider using apiserver_resource_objects instead] Number of stored objects at the time of last check split by kind. In case of a fetching error, the value will be -1.
    • STABLE
    • Gauge
      • -
    • resource
    +
  • resource
  • 1.34.0
    • apiserver_storage_size_bytes
    Size of the storage database file physically allocated in bytes.
    @@ -242,13 +242,6 @@ Stable metrics observe strict API contracts and no labels can be added or remove
  • Histogram
    • -
    scheduler_pod_scheduling_duration_seconds
    -
    E2e latency for a pod being scheduled which may include multiple scheduling attempts.
    -
      -
    • STABLE
      • -
    • Histogram
      • -
    • attempts
    • 1.29.0
    -
    scheduler_preemption_attempts_total
    Total preemption attempts in the cluster till now
      @@ -291,6 +284,34 @@ Stable metrics observe strict API contracts and no labels can be added or remove Beta metrics observe a looser API contract than its stable counterparts. No labels can be removed from beta metrics during their lifetime, however, labels can be added while the metric is in the beta stage. This offers the assurance that beta metrics will honor existing dashboards and alerts, while allowing for amendments in the future.
      +
      apiserver_authentication_config_controller_automatic_reload_last_timestamp_seconds
      +
      Timestamp of the last automatic reload of authentication configuration split by status and apiserver identity.
      +
        +
      • BETA
        • +
      • Gauge
        • +
      • apiserver_id_hashstatus
      +
      +
      apiserver_authentication_config_controller_automatic_reloads_total
      +
      Total number of automatic reloads of authentication configuration split by status and apiserver identity.
      +
        +
      • BETA
        • +
      • Counter
        • +
      • apiserver_id_hashstatus
      +
      +
      apiserver_authorization_config_controller_automatic_reload_last_timestamp_seconds
      +
      Timestamp of the last automatic reload of authorization configuration split by status and apiserver identity.
      +
        +
      • BETA
        • +
      • Gauge
        • +
      • apiserver_id_hashstatus
      +
      +
      apiserver_authorization_config_controller_automatic_reloads_total
      +
      Total number of automatic reloads of authorization configuration split by status and apiserver identity.
      +
        +
      • BETA
        • +
      • Counter
        • +
      • apiserver_id_hashstatus
      +
      apiserver_cel_compilation_duration_seconds
      CEL compilation time in seconds.
        @@ -368,6 +389,20 @@ Beta metrics observe a looser API contract than its stable counterparts. No labe
      • Counter
      • enforcement_actionerror_typepolicypolicy_binding
      +
      apiserver_validation_declarative_validation_mismatch_total
      +
      Number of times declarative validation results differed from handwritten validation results for core types.
      +
        +
      • BETA
        • +
      • Counter
        • +
      +
      +
      apiserver_validation_declarative_validation_panic_total
      +
      Number of times declarative validation has panicked during validation.
      +
        +
      • BETA
        • +
      • Counter
        • +
      +
      disabled_metrics_total
      The count of disabled metrics.
        @@ -389,6 +424,13 @@ Beta metrics observe a looser API contract than its stable counterparts. No labe
      • Gauge
      • namestage
      +
      prober_probe_total
      +
      Cumulative number of a liveness, readiness or startup probe for a container by result.
      +
        +
      • BETA
        • +
      • Counter
        • +
      • containernamespacepodpod_uidprobe_typeresult
      +
      registered_metrics_total
      The count of registered metrics broken by stability level and deprecation version.
        @@ -543,19 +585,12 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
      • Counter
      -
      apiserver_authentication_config_controller_automatic_reload_last_timestamp_seconds
      -
      Timestamp of the last automatic reload of authentication configuration split by status and apiserver identity.
      +
      apiserver_authentication_config_controller_last_config_info
      +
      Information about the last applied authentication configuration with hash as label, split by apiserver identity.
      • ALPHA
        • -
      • Gauge
        • -
      • apiserver_id_hashstatus
      -
      -
      apiserver_authentication_config_controller_automatic_reloads_total
      -
      Total number of automatic reloads of authentication configuration split by status and apiserver identity.
      -
        -
      • ALPHA
        • -
      • Counter
        • -
      • apiserver_id_hashstatus
      +
    • Custom
      • +
    • apiserver_id_hashhash
    apiserver_authentication_jwt_authenticator_latency_seconds
    Latency of jwt authentication operations in seconds. This is the time spent authenticating a token for cache miss only (i.e. when the token is not found in the cache).
    @@ -564,19 +599,12 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
  • Histogram
  • jwt_issuer_hashresult
    • -
    apiserver_authorization_config_controller_automatic_reload_last_timestamp_seconds
    -
    Timestamp of the last automatic reload of authorization configuration split by status and apiserver identity.
    +
    apiserver_authorization_config_controller_last_config_info
    +
    Information about the last applied authorization configuration with hash as label, split by apiserver identity.
    • ALPHA
      • -
    • Gauge
      • -
    • apiserver_id_hashstatus
    -
    -
    apiserver_authorization_config_controller_automatic_reloads_total
    -
    Total number of automatic reloads of authorization configuration split by status and apiserver identity.
    -
      -
    • ALPHA
      • -
    • Counter
      • -
    • apiserver_id_hashstatus
    +
  • Custom
    • +
  • apiserver_id_hashhash
    • apiserver_authorization_decisions_total
    Total number of terminal decisions made by an authorizer split by authorizer type, name, and decision.
    @@ -632,21 +660,21 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • ALPHA
    • Counter
      • -
    • indexresource_prefix
    +
  • groupindexresource
    • apiserver_cache_list_returned_objects_total
    Number of objects returned for a LIST request from watch cache
    • ALPHA
    • Counter
      • -
    • resource_prefix
    +
  • groupresource
    • apiserver_cache_list_total
    Number of LIST requests served from watch cache
    • ALPHA
    • Counter
      • -
    • indexresource_prefix
    +
  • groupindexresource
    • apiserver_certificates_registry_csr_honored_duration_total
    Total number of issued CSRs with a requested duration that was honored, sliced by signer (only kubernetes.io signer names are specifically identified)
    @@ -760,13 +788,6 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
  • Counter
  • protocoltransport
    • -
    apiserver_encryption_config_controller_automatic_reload_failures_total
    -
    Total number of failed automatic reloads of encryption configuration split by apiserver identity.
    -
      -
    • ALPHA
      • -
    • Counter
      • -
    • apiserver_id_hash
    • 1.30.0
    -
    apiserver_encryption_config_controller_automatic_reload_last_timestamp_seconds
    Timestamp of the last successful or failed automatic reload of encryption configuration split by apiserver identity.
      @@ -774,13 +795,6 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • Gauge
    • apiserver_id_hashstatus
    -
    apiserver_encryption_config_controller_automatic_reload_success_total
    -
    Total number of successful automatic reloads of encryption configuration split by apiserver identity.
    -
      -
    • ALPHA
      • -
    • Counter
      • -
    • apiserver_id_hash
    • 1.30.0
    -
    apiserver_encryption_config_controller_automatic_reloads_total
    Total number of reload successes and failures of encryption configuration split by apiserver identity.
      @@ -788,6 +802,13 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • Counter
    • apiserver_id_hashstatus
    +
    apiserver_encryption_config_controller_last_config_info
    +
    Information about the last applied encryption configuration with hash as label, split by apiserver identity.
    +
      +
    • ALPHA
      • +
    • Custom
      • +
    • apiserver_id_hashhash
    +
    apiserver_envelope_encryption_dek_cache_fill_percent
    Percent of the cache slots currently occupied by cached DEKs.
      @@ -1073,7 +1094,7 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
      • ALPHA
      • Counter
        • -
      • resource
      +
    • groupresource
    apiserver_kube_aggregator_x509_insecure_sha1_total
    Counts the number of requests to servers with insecure SHA1 signatures in their serving certificate OR the number of connection failures due to the insecure SHA1 signatures (either/or, based on the runtime environment)
    @@ -1089,6 +1110,20 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
  • Counter
    • +
    apiserver_mutating_admission_policy_check_duration_seconds
    +
    Mutation admission latency for individual mutation expressions in seconds, labeled by policy and binding.
    +
      +
    • ALPHA
      • +
    • Histogram
      • +
    • error_typepolicypolicy_binding
    +
    +
    apiserver_mutating_admission_policy_check_total
    +
    Mutation admission policy check total, labeled by policy and further identified by binding.
    +
      +
    • ALPHA
      • +
    • Counter
      • +
    • error_typepolicypolicy_binding
    +
    apiserver_nodeport_repair_port_errors_total
    Number of errors detected on ports by the repair loop broken down by type of error: leak, repair, full, outOfRange, duplicate, unknown
      @@ -1115,7 +1150,7 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
      • ALPHA
      • Histogram
        • -
      • resourceverb
      +
    • groupresourceverb
    apiserver_request_filter_duration_seconds
    Request filter latency distribution in seconds, for each filter type
    @@ -1166,12 +1201,33 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
  • Counter
  • code
    • +
    apiserver_resource_objects
    +
    Number of stored objects at the time of last check split by kind. In case of a fetching error, the value will be -1.
    +
      +
    • ALPHA
      • +
    • Gauge
      • +
    • groupresource
    +
    +
    apiserver_resource_size_estimate_bytes
    +
    Estimated size of stored objects in database. Estimate is based on sum of last observed sizes of serialized objects. In case of a fetching error, the value will be -1.
    +
      +
    • ALPHA
      • +
    • Gauge
      • +
    • groupresource
    +
    apiserver_selfrequest_total
    Counter of apiserver self-requests broken out for each verb, API resource and subresource.
    • ALPHA
    • Counter
      • -
    • resourcesubresourceverb
    +
  • groupresourcesubresourceverb
    • +
    +
    apiserver_storage_consistency_checks_total
    +
    Counter for status of consistency checks between etcd and watch cache
    +
      +
    • ALPHA
      • +
    • Counter
      • +
    • groupresourcestatus
    apiserver_storage_data_key_generation_duration_seconds
    Latencies in seconds of data encryption key(DEK) generation operations.
    @@ -1199,7 +1255,7 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • ALPHA
    • Counter
      • -
    • resource
    +
  • groupresource
    • apiserver_storage_envelope_transformation_cache_misses_total
    Total number of cache misses while accessing key decryption key(KEK).
    @@ -1213,35 +1269,35 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • ALPHA
    • Counter
      • -
    • resource
    +
  • groupresource
    • apiserver_storage_list_evaluated_objects_total
    Number of objects tested in the course of serving a LIST request from storage
    • ALPHA
    • Counter
      • -
    • resource
    +
  • groupresource
    • apiserver_storage_list_fetched_objects_total
    Number of objects read from storage in the course of serving a LIST request
    • ALPHA
    • Counter
      • -
    • resource
    +
  • groupresource
    • apiserver_storage_list_returned_objects_total
    Number of objects returned for a LIST request from storage
    • ALPHA
    • Counter
      • -
    • resource
    +
  • groupresource
    • apiserver_storage_list_total
    Number of LIST requests served from storage
    • ALPHA
    • Counter
      • -
    • resource
    +
  • groupresource
    • apiserver_storage_transformation_duration_seconds
    Latencies in seconds of value transformation operations.
    @@ -1276,7 +1332,7 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • ALPHA
    • Counter
      • -
    • resource
    +
  • groupresource
    • apiserver_tls_handshake_errors_total
    Number of requests dropped with 'TLS handshake error from' error
    @@ -1290,56 +1346,56 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • ALPHA
    • Counter
      • -
    • fallbackresourcesuccess
    +
  • fallbackgroupresourcesuccess
    • apiserver_watch_cache_events_dispatched_total
    Counter of events dispatched in watch cache broken by resource type.
    • ALPHA
    • Counter
      • -
    • resource
    +
  • groupresource
    • apiserver_watch_cache_events_received_total
    Counter of events received in watch cache broken by resource type.
    • ALPHA
    • Counter
      • -
    • resource
    +
  • groupresource
    • apiserver_watch_cache_initializations_total
    Counter of watch cache initializations broken by resource type.
    • ALPHA
    • Counter
      • -
    • resource
    +
  • groupresource
    • apiserver_watch_cache_read_wait_seconds
    Histogram of time spent waiting for a watch cache to become fresh.
    • ALPHA
    • Histogram
      • -
    • resource
    +
  • groupresource
    • apiserver_watch_cache_resource_version
    Current resource version of watch cache broken by resource type.
    • ALPHA
    • Gauge
      • -
    • resource
    +
  • groupresource
    • apiserver_watch_events_sizes
    Watch event size distribution in bytes
    • ALPHA
    • Histogram
      • -
    • groupkindversion
    +
  • groupresourceversion
    • apiserver_watch_events_total
    Number of events sent in watch clients
    • ALPHA
    • Counter
      • -
    • groupkindversion
    +
  • groupresourceversion
    • apiserver_watch_list_duration_seconds
    Response latency distribution in seconds for watch list requests broken by group, version, resource and scope.
    @@ -1467,6 +1523,13 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
  • Counter
  • code
    • +
    container_swap_limit_bytes
    +
    Current amount of the container swap limit in bytes. Reported only on non-windows systems
    +
      +
    • ALPHA
      • +
    • Custom
      • +
    • containerpodnamespace
    +
    container_swap_usage_bytes
    Current amount of the container swap usage in bytes. Reported only on non-windows systems
      @@ -1481,6 +1544,20 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • Histogram
    • driver_namegrpc_status_codemethod_namemigrated
    +
    device_taint_eviction_controller_pod_deletion_duration_seconds
    +
    Latency, in seconds, between the time when a device taint effect has been activated and a Pod's deletion via DeviceTaintEvictionController.
    +
      +
    • ALPHA
      • +
    • Histogram
      • +
    +
    +
    device_taint_eviction_controller_pod_deletions_total
    +
    Total number of Pods deleted by DeviceTaintEvictionController since its start.
    +
      +
    • ALPHA
      • +
    • Counter
      • +
    +
    dra_grpc_operations_duration_seconds
    Duration in seconds of the DRA gRPC operations
      @@ -1495,6 +1572,13 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • Histogram
    • is_erroroperation_name
    +
    dra_resource_claims_in_use
    +
    The number of ResourceClaims that are currently in use on the node, by driver name (driver_name label value) and across all drivers (special value for driver_name). Note that the sum of all by-driver counts is not the total number of in-use ResourceClaims because the same ResourceClaim might use devices from different drivers. Instead, use the count for the driver_name.
    +
      +
    • ALPHA
      • +
    • Custom
      • +
    • driver_name
    +
    endpoint_slice_controller_changes
    Number of EndpointSlice changes
      @@ -1640,7 +1724,7 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
      • ALPHA
      • Gauge
        • -
      • resource
      +
    • groupresource
    etcd_lease_object_counts
    Number of objects attached to a single etcd lease.
    @@ -1654,21 +1738,21 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • ALPHA
    • Histogram
      • -
    • operationtype
    +
  • groupoperationresource
    • etcd_request_errors_total
    Etcd failed request counts for each operation and object type.
    • ALPHA
    • Counter
      • -
    • operationtype
    +
  • groupoperationresource
    • etcd_requests_total
    Etcd request counts for each operation and object type.
    • ALPHA
    • Counter
      • -
    • operationtype
    +
  • groupoperationresource
    • etcd_version_info
    Etcd server's binary version
    @@ -1922,6 +2006,13 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
  • Counter
  • boundaryscope
    • +
    kubelet_container_aligned_compute_resources_failure_count
    +
    Cumulative number of failures to allocate aligned compute resources to containers by alignment type.
    +
      +
    • ALPHA
      • +
    • Counter
      • +
    • boundaryscope
    +
    kubelet_container_log_filesystem_used_bytes
    Bytes used by the container's logs on the filesystem.
      @@ -1929,6 +2020,13 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • Custom
    • uidnamespacepodcontainer
    +
    kubelet_container_requested_resizes_total
    +
    Number of requested resizes, counted at the container level. Different resources on the same container are counted separately. The 'requirement' label refers to 'memory' or 'limits'; the 'operation' label can be one of 'add', 'remove', 'increase' or 'decrease'.
    +
      +
    • ALPHA
      • +
    • Counter
      • +
    • operationrequirementresource
    +
    kubelet_containers_per_pod_count
    The number of containers per pod.
      @@ -1936,6 +2034,13 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • Histogram
    +
    kubelet_cpu_manager_allocation_per_numa
    +
    Number of CPUs allocated per NUMA node
    +
      +
    • ALPHA
      • +
    • Gauge
      • +
    • numa_node
    +
    kubelet_cpu_manager_exclusive_cpu_allocation_count
    The total number of CPUs exclusively allocated to containers running on this node
      @@ -1964,6 +2069,13 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • Gauge
    +
    kubelet_credential_provider_config_info
    +
    Information about the last applied credential provider configuration with hash as label
    +
      +
    • ALPHA
      • +
    • Custom
      • +
    • hash
    +
    kubelet_credential_provider_plugin_duration
    Duration of execution in seconds for credential provider plugin
      @@ -1971,13 +2083,20 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • Histogram
    • plugin_name
    -
    kubelet_credential_provider_plugin_errors
    +
    kubelet_credential_provider_plugin_errors_total
    Number of errors from credential provider plugin
    • ALPHA
    • Counter
    • plugin_name
    +
    kubelet_cri_losing_support
    +
    the Kubernetes version that the currently running CRI implementation will lose support on if not upgraded.
    +
      +
    • ALPHA
      • +
    • Gauge
      • +
    • version
    +
    kubelet_desired_pods
    The number of pods the kubelet is being instructed to run. static is true if the pod is not from the apiserver.
      @@ -2083,6 +2202,27 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • Histogram
    • image_size_in_bytes
    +
    kubelet_image_volume_mounted_errors_total
    +
    Number of failed image volume mounts.
    +
      +
    • ALPHA
      • +
    • Counter
      • +
    +
    +
    kubelet_image_volume_mounted_succeed_total
    +
    Number of successful image volume mounts.
    +
      +
    • ALPHA
      • +
    • Counter
      • +
    +
    +
    kubelet_image_volume_requested_total
    +
    Number of requested image volumes.
    +
      +
    • ALPHA
      • +
    • Counter
      • +
    +
    kubelet_lifecycle_handler_http_fallbacks_total
    The number of times lifecycle handlers successfully fell back to http from https.
      @@ -2209,6 +2349,41 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • Histogram
    +
    kubelet_pod_deferred_accepted_resizes_total
    +
    Cumulative number of resizes that were accepted after being deferred.
    +
      +
    • ALPHA
      • +
    • Counter
      • +
    • retry_trigger
    +
    +
    kubelet_pod_in_progress_resizes
    +
    Number of in-progress resizes for pods.
    +
      +
    • ALPHA
      • +
    • Gauge
      • +
    +
    +
    kubelet_pod_infeasible_resizes_total
    +
    Number of infeasible resizes for pods.
    +
      +
    • ALPHA
      • +
    • Counter
      • +
    • reason_detail
    +
    +
    kubelet_pod_pending_resizes
    +
    Number of pending resizes for pods.
    +
      +
    • ALPHA
      • +
    • Gauge
      • +
    • reason
    +
    +
    kubelet_pod_resize_duration_milliseconds
    +
    Duration in milliseconds to actuate a pod resize
    +
      +
    • ALPHA
      • +
    • Histogram
      • +
    • success
    +
    kubelet_pod_resources_endpoint_errors_get
    Number of requests to the PodResource Get endpoint which returned error. Broken down by server api version.
      @@ -2419,6 +2594,20 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • Counter
    +
    kubelet_started_user_namespaced_pods_errors_total
    +
    Cumulative number of errors when starting pods with user namespaces. This metric will only be collected on Linux.
    +
      +
    • ALPHA
      • +
    • Counter
      • +
    +
    +
    kubelet_started_user_namespaced_pods_total
    +
    Cumulative number of pods with user namespaces started. This metric will only be collected on Linux.
    +
      +
    • ALPHA
      • +
    • Counter
      • +
    +
    kubelet_topology_manager_admission_duration_ms
    Duration in milliseconds to serve a pod admission request.
      @@ -2503,6 +2692,20 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • Gauge
    • configlifecyclestatic
    +
    kubeproxy_conntrack_reconciler_deleted_entries_total
    +
    Cumulative conntrack flows deleted by conntrack reconciler
    +
      +
    • ALPHA
      • +
    • Counter
      • +
    • ip_family
    +
    +
    kubeproxy_conntrack_reconciler_sync_duration_seconds
    +
    ReconcileConntrackFlowsLatency latency in seconds
    +
      +
    • ALPHA
      • +
    • Histogram
      • +
    • ip_family
    +
    kubeproxy_iptables_ct_state_invalid_dropped_packets_total
    packets dropped by iptables to work around conntrack problems
      @@ -2522,7 +2725,7 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
      • ALPHA
      • Histogram
        • -
      +
    • ip_family
    kubeproxy_proxy_healthz_total
    Cumulative proxy healthz HTTP status
    @@ -2543,21 +2746,21 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • ALPHA
    • Histogram
      • -
    +
  • ip_family
    • kubeproxy_sync_partial_proxy_rules_duration_seconds
    SyncProxyRules latency in seconds for partial resyncs
    • ALPHA
    • Histogram
      • -
    +
  • ip_family
    • kubeproxy_sync_proxy_rules_duration_seconds
    SyncProxyRules latency in seconds
    • ALPHA
    • Histogram
      • -
    +
  • ip_family
    • kubeproxy_sync_proxy_rules_endpoint_changes_pending
    Pending proxy rules Endpoint changes
    @@ -2578,63 +2781,63 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • ALPHA
    • Gauge
      • -
    • table
    +
  • ip_familytable
    • kubeproxy_sync_proxy_rules_iptables_partial_restore_failures_total
    Cumulative proxy iptables partial restore failures
    • ALPHA
    • Counter
      • -
    +
  • ip_family
    • kubeproxy_sync_proxy_rules_iptables_restore_failures_total
    Cumulative proxy iptables restore failures
    • ALPHA
    • Counter
      • -
    +
  • ip_family
    • kubeproxy_sync_proxy_rules_iptables_total
    Total number of iptables rules owned by kube-proxy
    • ALPHA
    • Gauge
      • -
    • table
    +
  • ip_familytable
    • kubeproxy_sync_proxy_rules_last_queued_timestamp_seconds
    The last time a sync of proxy rules was queued
    • ALPHA
    • Gauge
      • -
    +
  • ip_family
    • kubeproxy_sync_proxy_rules_last_timestamp_seconds
    The last time proxy rules were successfully synced
    • ALPHA
    • Gauge
      • -
    +
  • ip_family
    • kubeproxy_sync_proxy_rules_nftables_cleanup_failures_total
    Cumulative proxy nftables cleanup failures
    • ALPHA
    • Counter
      • -
    +
  • ip_family
    • kubeproxy_sync_proxy_rules_nftables_sync_failures_total
    Cumulative proxy nftables sync failures
    • ALPHA
    • Counter
      • -
    +
  • ip_family
    • kubeproxy_sync_proxy_rules_no_local_endpoints_total
    Number of services with a Local traffic policy and no endpoints
    • ALPHA
    • Gauge
      • -
    • traffic_policy
    +
  • ip_familytraffic_policy
    • kubeproxy_sync_proxy_rules_service_changes_pending
    Pending proxy rules Service changes
    @@ -2825,13 +3028,6 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
  • Histogram
  • containernamespacepodprobe_type
    • -
    prober_probe_total
    -
    Cumulative number of a liveness, readiness or startup probe for a container by result.
    -
      -
    • ALPHA
      • -
    • Counter
      • -
    • containernamespacepodpod_uidprobe_typeresult
    -
    pv_collector_bound_pv_count
    Gauge measuring number of persistent volume currently bound
      @@ -2888,33 +3084,19 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • Histogram
    -
    resourceclaim_controller_allocated_resource_claims
    -
    Number of allocated ResourceClaims
    -
      -
    • ALPHA
      • -
    • Gauge
      • -
    -
    -
    resourceclaim_controller_create_attempts_total
    -
    Number of ResourceClaims creation requests
    +
    resourceclaim_controller_creates_total
    +
    Number of ResourceClaims creation requests, categorized by creation status and admin access
    • ALPHA
    • Counter
      • -
    -
    -
    resourceclaim_controller_create_failures_total
    -
    Number of ResourceClaims creation request failures
    -
      -
    • ALPHA
      • -
    • Counter
      • -
    +
  • admin_accessstatus
    • resourceclaim_controller_resource_claims
    -
    Number of ResourceClaims
    +
    Number of ResourceClaims, categorized by allocation status and admin access
    • ALPHA
      • -
    • Gauge
      • -
    +
  • Custom
    • +
  • allocatedadmin_access
    • rest_client_dns_resolution_duration_seconds
    DNS resolver latency in seconds. Broken down by host.
    @@ -3035,6 +3217,27 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
  • Gauge
  • managername
    • +
    scheduler_async_api_call_execution_duration_seconds
    +
    Duration in seconds for executing API call in the async dispatcher.
    +
      +
    • ALPHA
      • +
    • Histogram
      • +
    • call_typeresult
    +
    +
    scheduler_async_api_call_execution_total
    +
    Total number of API calls executed by the async dispatcher.
    +
      +
    • ALPHA
      • +
    • Counter
      • +
    • call_typeresult
    +
    +
    scheduler_cache_size
    +
    Number of nodes, pods, and assumed (bound) pods in the scheduler cache.
    +
      +
    • ALPHA
      • +
    • Gauge
      • +
    • type
    +
    scheduler_event_handling_duration_seconds
    Event handling latency in seconds.
      @@ -3056,6 +3259,13 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • Gauge
    • event
    +
    scheduler_pending_async_api_calls
    +
    Number of API calls currently pending in the async queue.
    +
      +
    • ALPHA
      • +
    • Gauge
      • +
    • call_type
    +
    scheduler_permit_wait_duration_seconds
    Duration of waiting on permit.
      @@ -3098,13 +3308,6 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • Histogram
    • eventhintplugin
    -
    scheduler_scheduler_cache_size
    -
    Number of nodes, pods, and assumed (bound) pods in the scheduler cache.
    -
      -
    • ALPHA
      • -
    • Gauge
      • -
    • type
    -
    scheduler_scheduling_algorithm_duration_seconds
    Scheduling algorithm latency in seconds
      @@ -3252,6 +3455,13 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
    • Histogram
    +
    version_info
    +
    Provides the compatibility version info of the component. The component label is the name of the component, usually kube, but is relevant for aggregated-apiservers.
    +
      +
    • ALPHA
      • +
    • Gauge
      • +
    • binarycomponentemulationmin_compat
    +
    volume_manager_selinux_container_errors_total
    Number of errors when kubelet cannot compute SELinux context for a container. Kubelet can't start such a Pod then and it will retry, therefore value of this metric may not represent the actual nr. of containers.
      @@ -3327,21 +3537,21 @@ Alpha metrics do not have any API guarantees. These metrics must be used at your
      • ALPHA
      • Gauge
        • -
      • resource
      +
    • groupresource
    watch_cache_capacity_decrease_total
    Total number of watch cache capacity decrease events broken by resource type.
    • ALPHA
    • Counter
      • -
    • resource
    +
  • groupresource
    • watch_cache_capacity_increase_total
    Total number of watch cache capacity increase events broken by resource type.
    • ALPHA
    • Counter
      • -
    • resource
    +
  • groupresource
    • workqueue_adds_total
    Total number of adds handled by workqueue