diff --git a/test/instrumentation/decode_metric.go b/test/instrumentation/decode_metric.go index 7f2821d1604..3946b9f3b3d 100644 --- a/test/instrumentation/decode_metric.go +++ b/test/instrumentation/decode_metric.go @@ -126,17 +126,17 @@ func (c *metricDecoder) decodeDesc(ce *ast.CallExpr) (metric, error) { m := &metric{} name, err := c.decodeString(ce.Args[0]) if err != nil { - return *m, newDecodeErrorf(ce, "can't decode string") + return *m, newDecodeErrorf(ce, errorDecodingString) } m.Name = *name help, err := c.decodeString(ce.Args[1]) if err != nil { - return *m, newDecodeErrorf(ce, "can't decode string") + return *m, newDecodeErrorf(ce, errorDecodingString) } m.Help = *help labels, err := c.decodeLabels(ce.Args[2]) if err != nil { - return *m, newDecodeErrorf(ce, "can't decode labels") + return *m, newDecodeErrorf(ce, errorDecodingLabels) } m.Labels = labels cLabels, err := c.decodeConstLabels(ce.Args[3]) @@ -153,7 +153,7 @@ func (c *metricDecoder) decodeDesc(ce *ast.CallExpr) (metric, error) { } deprecatedVersion, err := c.decodeString(ce.Args[5]) if err != nil { - return *m, newDecodeErrorf(ce, "can't decode string") + return *m, newDecodeErrorf(ce, errorDecodingString) } if deprecatedVersion != nil { m.DeprecatedVersion = *deprecatedVersion @@ -250,7 +250,7 @@ func (c *metricDecoder) decodeString(expr ast.Expr) (*string, error) { } return &value, nil } - return nil, fmt.Errorf("can't decode string") + return nil, newDecodeErrorf(expr, errorDecodingString) } func (c *metricDecoder) decodeMetricVec(call *ast.CallExpr) (metric, error) { @@ -307,11 +307,11 @@ func (c *metricDecoder) decodeLabels(expr ast.Expr) ([]string, error) { } variableExpr, found := c.variables[e.Name] if !found { - return nil, newDecodeErrorf(expr, "couldn't find variable for labels") + return nil, newDecodeErrorf(expr, errorFindingVariableForLabels) } cl2, ok := variableExpr.(*ast.CompositeLit) if !ok { - return nil, newDecodeErrorf(expr, "couldn't interpret variable for labels") + return nil, newDecodeErrorf(expr, errorFindingVariableForLabels) } cl = cl2 } @@ -435,7 +435,7 @@ func (c *metricDecoder) decodeBuckets(expr ast.Expr) ([]float64, error) { return float64s, err2 } default: - return nil, newDecodeErrorf(v, "couldn't find variable for bucket") + return nil, newDecodeErrorf(v, errorFindingVariableForBuckets) } case *ast.CompositeLit: @@ -478,7 +478,7 @@ func (c *metricDecoder) decodeBucketFunctionCall(v *ast.CallExpr) ([]float64, er } firstArg, secondArg, thirdArg, err := decodeBucketArguments(v2) if err != nil { - return nil, err, true + return nil, newDecodeErrorf(v, errBuckets), true } switch functionName { case "LinearBuckets": @@ -520,7 +520,7 @@ func (c *metricDecoder) decodeBucketFunctionCall(v *ast.CallExpr) ([]float64, er case *ast.CompositeLit: fs, err := decodeListOfFloats(argExpr, argExpr.Elts) if err != nil { - return nil, err, true + return nil, newDecodeErrorf(v, errBuckets), true } merged = append(merged, fs...) case *ast.CallExpr: @@ -536,7 +536,7 @@ func (c *metricDecoder) decodeBucketFunctionCall(v *ast.CallExpr) ([]float64, er } firstArg, secondArg, thirdArg, err := decodeBucketArguments(argExpr) if err != nil { - return nil, err, true + return nil, newDecodeErrorf(v, errBuckets), true } switch functionName { case "LinearBuckets": @@ -583,11 +583,11 @@ func (c *metricDecoder) decodeUint32(expr ast.Expr) (uint32, error) { if ok && importName.String() == c.kubeMetricsImportName { if variableName == "DefAgeBuckets" { // hardcode this for now - return 5, nil + return metrics.DefAgeBuckets, nil } if variableName == "DefBufCap" { // hardcode this for now - return 500, nil + return metrics.DefBufCap, nil } } case *ast.CallExpr: @@ -617,9 +617,9 @@ func (c *metricDecoder) decodeInt64(expr ast.Expr) (int64, error) { importName, ok := v.X.(*ast.Ident) if ok && importName.String() == c.kubeMetricsImportName { if variableName == "DefMaxAge" { - // hardcode this for now. This is a duration but we'll output it as + // hardcode this for now. This is a duration, but we'll output it as // an int64 representing nanoseconds. - return 1000 * 1000 * 1000 * 60 * 10, nil + return int64(metrics.DefMaxAge), nil } } case *ast.Ident: diff --git a/test/instrumentation/doc.go b/test/instrumentation/doc.go index 39f5916bda8..712e115ddbd 100644 --- a/test/instrumentation/doc.go +++ b/test/instrumentation/doc.go @@ -25,5 +25,17 @@ all files in the Kubernetes code base to: Due to the dynamic nature of how metrics can be written, we only support the subset of metrics which can actually be parsed. If a metric cannot be parsed, it must be delegated to the stability class `Internal`, which will exempt the metric from static analysis. + +The entrypoint to this package is defined in a shell script (i.e. stability-utils.sh) which has +the logic for feeding file names as arguments into the program. The logic of this program is as +follows: + + - parse all files fed in, keeping track of: + - the function and struct pointers which correspond to prometheus metric definitions. + - consts/variable we encounter, so that we can use these to resolve values in metric definitions + - then, iterate over the function and struct pointers, resolving attributes to concrete metric values + - then, using our collected and resolved metric definitions, output (depending on the mode): + - a yaml file corresponding to all stable metrics + - a documentation file corresponding to all parseable metrics in the Kubernetes codebase */ package main diff --git a/test/instrumentation/documentation/documentation-list.yaml b/test/instrumentation/documentation/documentation-list.yaml index ed86dca6993..dad1138b103 100644 --- a/test/instrumentation/documentation/documentation-list.yaml +++ b/test/instrumentation/documentation/documentation-list.yaml @@ -12,23 +12,6 @@ certificate is invalid or unused, the value will be +INF. type: Gauge stabilityLevel: ALPHA -- name: cronjob_job_creation_skew_duration_seconds - subsystem: cronjob_controller - help: Time between when a cronjob is scheduled to be run, and when the corresponding - job is created - type: Histogram - stabilityLevel: ALPHA - buckets: - - 1 - - 2 - - 4 - - 8 - - 16 - - 32 - - 64 - - 128 - - 256 - - 512 - name: changes subsystem: endpoint_slice_controller help: Number of EndpointSlice changes @@ -270,6 +253,23 @@ stabilityLevel: ALPHA labels: - code +- name: job_creation_skew_duration_seconds + subsystem: cronjob_controller + help: Time between when a cronjob is scheduled to be run, and when the corresponding + job is created + type: Histogram + stabilityLevel: STABLE + buckets: + - 1 + - 2 + - 4 + - 8 + - 16 + - 32 + - 64 + - 128 + - 256 + - 512 - name: attachdetach_controller_forced_detaches help: Number of times the A/D Controller performed a forced detach type: Counter @@ -281,57 +281,6 @@ labels: - plugin_name - state -- name: job_finished_total - subsystem: job_controller - help: The number of finished job - type: Counter - stabilityLevel: ALPHA - labels: - - completion_mode - - reason - - result -- name: job_pods_finished_total - subsystem: job_controller - help: The number of finished Pods that are fully tracked - type: Counter - stabilityLevel: ALPHA - labels: - - completion_mode - - result -- name: job_sync_duration_seconds - subsystem: job_controller - help: The time it took to sync a job - type: Histogram - stabilityLevel: ALPHA - labels: - - action - - completion_mode - - 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: job_sync_total - subsystem: job_controller - help: The number of job syncs - type: Counter - stabilityLevel: ALPHA - labels: - - action - - completion_mode - - result - name: pod_failures_handled_by_failure_policy_total subsystem: job_controller help: "`The number of failed Pods handled by failure policy with\n\t\t\trespect @@ -450,6 +399,18 @@ stabilityLevel: ALPHA labels: - clusterCIDR +- name: force_delete_pod_errors_total + subsystem: pod_gc_collector + help: Number of errors encountered when forcefully deleting the pods since the Pod + GC Controller started. + type: Counter + stabilityLevel: ALPHA +- name: force_delete_pods_total + subsystem: pod_gc_collector + help: Number of pods that are being forcefully deleted since the Pod GC Controller + started. + type: Counter + stabilityLevel: ALPHA - name: sorting_deletion_age_ratio subsystem: replicaset_controller help: The ratio of chosen deleted pod's ages to the current youngest pod's age (at @@ -493,6 +454,57 @@ - 204.8 - 409.6 - 819.2 +- name: job_pods_finished_total + subsystem: job_controller + help: The number of finished Pods that are fully tracked + type: Counter + stabilityLevel: STABLE + labels: + - completion_mode + - result +- name: job_sync_duration_seconds + subsystem: job_controller + help: The time it took to sync a job + type: Histogram + stabilityLevel: STABLE + labels: + - action + - completion_mode + - 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: job_syncs_total + subsystem: job_controller + help: The number of job syncs + type: Counter + stabilityLevel: STABLE + labels: + - action + - completion_mode + - result +- name: jobs_finished_total + subsystem: job_controller + help: The number of finished jobs + type: Counter + stabilityLevel: STABLE + labels: + - completion_mode + - reason + - result - name: evictions_total subsystem: node_collector help: Number of Node evictions that happened since current instance of NodeController @@ -758,48 +770,6 @@ help: Last graceful shutdown start time since unix epoch in seconds type: Gauge stabilityLevel: ALPHA -- name: http_inflight_requests - subsystem: kubelet - help: Number of the inflight http requests - type: Gauge - stabilityLevel: ALPHA - labels: - - long_running - - method - - path - - server_type -- name: http_requests_duration_seconds - subsystem: kubelet - help: Duration in seconds to serve http requests - type: Histogram - stabilityLevel: ALPHA - labels: - - long_running - - method - - path - - server_type - buckets: - - 0.005 - - 0.01 - - 0.025 - - 0.05 - - 0.1 - - 0.25 - - 0.5 - - 1 - - 2.5 - - 5 - - 10 -- name: http_requests_total - subsystem: kubelet - help: Number of the http requests received since the server started - type: Counter - stabilityLevel: ALPHA - labels: - - long_running - - method - - path - - server_type - name: lifecycle_handler_http_fallbacks_total subsystem: kubelet help: The number of times lifecycle handlers successfully fell back to http from @@ -922,6 +892,39 @@ - 2.5 - 5 - 10 +- name: pod_start_sli_duration_seconds + subsystem: kubelet + help: Duration in seconds to start a pod, excluding time to pull images and run + init containers, measured from pod creation timestamp to when all its containers + are reported as started and observed via watch + type: Histogram + stabilityLevel: ALPHA + buckets: + - 0.5 + - 1 + - 2 + - 3 + - 4 + - 5 + - 6 + - 8 + - 10 + - 20 + - 30 + - 45 + - 60 + - 120 + - 180 + - 240 + - 300 + - 360 + - 480 + - 600 + - 900 + - 1200 + - 1800 + - 2700 + - 3600 - name: pod_status_sync_duration_seconds subsystem: kubelet help: Duration in seconds to sync a pod status update. Measures time from detection @@ -1103,25 +1106,6 @@ help: Cumulative number of pods started type: Counter stabilityLevel: ALPHA -- name: volume_metric_collection_duration_seconds - subsystem: kubelet - help: Duration in seconds to calculate volume stats - type: Histogram - stabilityLevel: ALPHA - labels: - - metric_source - buckets: - - 0.005 - - 0.01 - - 0.025 - - 0.05 - - 0.1 - - 0.25 - - 0.5 - - 1 - - 2.5 - - 5 - - 10 - name: kubelet_volume_stats_available_bytes help: Number of available bytes in the volume type: Custom @@ -1228,24 +1212,67 @@ help: 1 if there was an error while getting container metrics, 0 otherwise type: Custom stabilityLevel: ALPHA -- 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: http_inflight_requests + subsystem: kubelet + help: Number of the inflight http requests + type: Gauge + stabilityLevel: ALPHA + labels: + - long_running + - method + - path + - server_type +- name: http_requests_duration_seconds + subsystem: kubelet + help: Duration in seconds to serve http requests + type: Histogram + stabilityLevel: ALPHA + labels: + - long_running + - method + - path + - server_type + buckets: + - 0.005 + - 0.01 + - 0.025 + - 0.05 + - 0.1 + - 0.25 + - 0.5 + - 1 + - 2.5 + - 5 + - 10 +- name: http_requests_total + subsystem: kubelet + help: Number of the http requests received since the server started 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 + - long_running + - method + - path + - server_type +- name: volume_metric_collection_duration_seconds + subsystem: kubelet + help: Duration in seconds to calculate volume stats + type: Histogram stabilityLevel: ALPHA labels: - - signerName + - metric_source + buckets: + - 0.005 + - 0.01 + - 0.025 + - 0.05 + - 0.1 + - 0.25 + - 0.5 + - 1 + - 2.5 + - 5 + - 10 - name: network_programming_duration_seconds subsystem: kubeproxy help: In Cluster Network Programming Latency in seconds @@ -1457,6 +1484,24 @@ 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: allocated_ips subsystem: clusterip_allocator namespace: kube_apiserver @@ -1723,7 +1768,9 @@ subsystem: scheduler help: Number of pending pods, by the queue type. 'active' means number of pods in activeQ; 'backoff' means number of pods in backoffQ; 'unschedulable' means number - of pods in unschedulablePods. + of pods in unschedulablePods that the scheduler attempted to schedule and failed; + 'gated' is the number of unschedulable pods that the scheduler never attempted + to schedule because they are gated. type: Gauge stabilityLevel: STABLE labels: @@ -1959,52 +2006,6 @@ - 4.096 - 8.192 - 16.384 -- name: check_duration_seconds - subsystem: validating_admission_policy - namespace: apiserver - help: Validation admission latency for individual validation expressions in seconds, - labeled by policy and param resource, further including binding, state and enforcement - action taken. - type: Histogram - stabilityLevel: ALPHA - labels: - - enforcement_action - - params - - policy - - policy_binding - - state - - validation_expression - buckets: - - 5e-07 - - 0.001 - - 0.01 - - 0.1 - - 1 -- name: check_total - subsystem: validating_admission_policy - namespace: apiserver - help: Validation admission policy check total, labeled by policy and param resource, - and further identified by binding, validation expression, enforcement action taken, - and state. - type: Counter - stabilityLevel: ALPHA - labels: - - enforcement_action - - params - - policy - - policy_binding - - state - - validation_expression -- name: definition_total - subsystem: validating_admission_policy - namespace: apiserver - help: Validation admission policy count total, labeled by state and enforcement - action. - type: Counter - stabilityLevel: ALPHA - labels: - - enforcement_action - - state - name: step_admission_duration_seconds_summary subsystem: admission namespace: apiserver @@ -2060,52 +2061,52 @@ - operation - rejected - type -- name: error_total - subsystem: apiserver_audit - help: Counter of audit events that failed to be audited properly. Plugin identifies - the plugin affected by the error. - type: Counter - stabilityLevel: ALPHA - labels: - - plugin -- name: event_total - subsystem: apiserver_audit - help: Counter of audit events generated and sent to the audit backend. - type: Counter - stabilityLevel: ALPHA -- name: level_total - subsystem: apiserver_audit - help: Counter of policy levels for audit events (1 per request). - type: Counter - stabilityLevel: ALPHA - labels: - - level -- name: requests_rejected_total - subsystem: apiserver_audit - help: Counter of apiserver requests rejected due to an error in audit logging backend. - type: Counter - stabilityLevel: ALPHA -- name: apiserver_delegated_authn_request_duration_seconds - help: Request latency in seconds. Broken down by status code. +- name: check_duration_seconds + subsystem: validating_admission_policy + namespace: apiserver + help: Validation admission latency for individual validation expressions in seconds, + labeled by policy and param resource, further including binding, state and enforcement + action taken. type: Histogram stabilityLevel: ALPHA labels: - - code + - enforcement_action + - params + - policy + - policy_binding + - state + - validation_expression buckets: - - 0.25 - - 0.5 - - 0.7 + - 5e-07 + - 0.001 + - 0.01 + - 0.1 - 1 - - 1.5 - - 3 - - 5 - - 10 -- name: apiserver_delegated_authn_request_total - help: Number of HTTP requests partitioned by status code. +- name: check_total + subsystem: validating_admission_policy + namespace: apiserver + help: Validation admission policy check total, labeled by policy and param resource, + and further identified by binding, validation expression, enforcement action taken, + and state. type: Counter stabilityLevel: ALPHA labels: - - code + - enforcement_action + - params + - policy + - policy_binding + - state + - validation_expression +- name: definition_total + subsystem: validating_admission_policy + namespace: apiserver + help: Validation admission policy count total, labeled by state and enforcement + action. + type: Counter + stabilityLevel: ALPHA + labels: + - enforcement_action + - state - name: controller_admission_duration_seconds subsystem: admission namespace: apiserver @@ -2162,6 +2163,31 @@ - 0.5 - 1 - 2.5 +- name: error_total + subsystem: apiserver_audit + help: Counter of audit events that failed to be audited properly. Plugin identifies + the plugin affected by the error. + type: Counter + stabilityLevel: ALPHA + labels: + - plugin +- name: event_total + subsystem: apiserver_audit + help: Counter of audit events generated and sent to the audit backend. + type: Counter + stabilityLevel: ALPHA +- name: level_total + subsystem: apiserver_audit + help: Counter of policy levels for audit events (1 per request). + type: Counter + stabilityLevel: ALPHA + labels: + - level +- name: requests_rejected_total + subsystem: apiserver_audit + help: Counter of apiserver requests rejected due to an error in audit logging backend. + type: Counter + stabilityLevel: ALPHA - name: compilation_duration_seconds subsystem: cel namespace: apiserver @@ -2202,6 +2228,27 @@ stabilityLevel: ALPHA labels: - request_kind +- name: apiserver_delegated_authn_request_duration_seconds + help: Request latency in seconds. Broken down by status code. + type: Histogram + stabilityLevel: ALPHA + labels: + - code + buckets: + - 0.25 + - 0.5 + - 0.7 + - 1 + - 1.5 + - 3 + - 5 + - 10 +- name: apiserver_delegated_authn_request_total + help: Number of HTTP requests partitioned by status code. + type: Counter + stabilityLevel: ALPHA + labels: + - code - name: apiserver_delegated_authz_request_duration_seconds help: Request latency in seconds. Broken down by status code. type: Histogram @@ -2223,45 +2270,6 @@ stabilityLevel: ALPHA labels: - code -- 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: request_aborts_total subsystem: apiserver help: Number of requests which apiserver aborted possibly due to a timeout, for @@ -2342,11 +2350,48 @@ labels: - source - status +- name: request_sli_duration_seconds + subsystem: apiserver + help: Response latency distribution (not counting webhook duration) in seconds for + each verb, group, version, resource, subresource, scope and component. + type: Histogram + stabilityLevel: ALPHA + labels: + - component + - group + - resource + - scope + - subresource + - verb + - version + buckets: + - 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: request_slo_duration_seconds subsystem: apiserver help: Response latency distribution (not counting webhook duration) in seconds for each verb, group, version, resource, subresource, scope and component. type: Histogram + deprecatedVersion: 1.27.0 stabilityLevel: ALPHA labels: - component @@ -2678,6 +2723,45 @@ labels: - index - resource_prefix +- 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: dek_cache_fill_percent subsystem: envelope_encryption namespace: apiserver @@ -2723,6 +2807,14 @@ labels: - flow_schema - priority_level +- name: current_limit_seats + subsystem: flowcontrol + namespace: apiserver + help: current derived number of execution seats available to each priority level + type: Gauge + stabilityLevel: ALPHA + labels: + - priority_level - name: current_r subsystem: flowcontrol namespace: apiserver @@ -2731,6 +2823,60 @@ stabilityLevel: ALPHA labels: - priority_level +- name: demand_seats + subsystem: flowcontrol + namespace: apiserver + help: Observations, at the end of every nanosecond, of (the number of seats each + priority level could use) / (nominal number of seats for that level) + type: TimingRatioHistogram + stabilityLevel: ALPHA + labels: + - priority_level + buckets: + - 0.2 + - 0.4 + - 0.6 + - 0.8 + - 1 + - 1.2 + - 1.4 + - 1.7 + - 2 + - 2.8 + - 4 + - 6 +- name: demand_seats_average + subsystem: flowcontrol + namespace: apiserver + help: Time-weighted average, over last adjustment period, of demand_seats + type: Gauge + stabilityLevel: ALPHA + labels: + - priority_level +- name: demand_seats_high_watermark + subsystem: flowcontrol + namespace: apiserver + help: High watermark, over last adjustment period, of demand_seats + type: Gauge + stabilityLevel: ALPHA + labels: + - priority_level +- name: demand_seats_smoothed + subsystem: flowcontrol + namespace: apiserver + help: Smoothed seat demands + type: Gauge + stabilityLevel: ALPHA + labels: + - priority_level +- name: demand_seats_stdev + subsystem: flowcontrol + namespace: apiserver + help: Time-weighted standard deviation, over last adjustment period, of demand_seats + type: Gauge + stabilityLevel: ALPHA + labels: + - priority_level - name: dispatch_r subsystem: flowcontrol namespace: apiserver @@ -2765,6 +2911,15 @@ stabilityLevel: ALPHA labels: - priority_level +- name: lower_limit_seats + subsystem: flowcontrol + namespace: apiserver + help: Configured lower bound on number of execution seats available to each priority + level + type: Gauge + stabilityLevel: ALPHA + labels: + - priority_level - name: next_discounted_s_bounds subsystem: flowcontrol namespace: apiserver @@ -2784,6 +2939,14 @@ labels: - bound - priority_level +- name: nominal_limit_seats + subsystem: flowcontrol + namespace: apiserver + help: Nominal number of execution seats configured for each priority level + type: Gauge + stabilityLevel: ALPHA + labels: + - priority_level - name: priority_level_request_utilization subsystem: flowcontrol namespace: apiserver @@ -2962,6 +3125,30 @@ - 5 - 10 - 30 +- name: seat_fair_frac + subsystem: flowcontrol + namespace: apiserver + help: Fair fraction of server's concurrency to allocate to each priority level that + can use it + type: Gauge + stabilityLevel: ALPHA +- name: target_seats + subsystem: flowcontrol + namespace: apiserver + help: Seat allocation targets + type: Gauge + stabilityLevel: ALPHA + labels: + - priority_level +- name: upper_limit_seats + subsystem: flowcontrol + namespace: apiserver + help: Configured upper bound on number of execution seats available to each priority + level + type: Gauge + stabilityLevel: ALPHA + labels: + - priority_level - name: watch_count_samples subsystem: flowcontrol namespace: apiserver @@ -3516,22 +3703,6 @@ - 0.1 - 1 - 10 -- name: x509_insecure_sha1_total - subsystem: kube_aggregator - namespace: apiserver - help: 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) - type: Counter - stabilityLevel: ALPHA -- name: x509_missing_san_total - subsystem: kube_aggregator - namespace: apiserver - help: Counts the number of requests to servers missing SAN extension in their serving - certificate OR the number of connection failures due to the lack of x509 certificate - SAN extension missing (either/or, based on the runtime environment) - type: Counter - stabilityLevel: ALPHA - name: aggregator_openapi_v2_regeneration_count help: Counter of OpenAPI v2 spec regeneration count broken down by causing APIService name and reason. @@ -3561,6 +3732,22 @@ labels: - name - reason +- name: x509_insecure_sha1_total + subsystem: kube_aggregator + namespace: apiserver + help: 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) + type: Counter + stabilityLevel: ALPHA +- name: x509_missing_san_total + subsystem: kube_aggregator + namespace: apiserver + help: Counts the number of requests to servers missing SAN extension in their serving + certificate OR the number of connection failures due to the lack of x509 certificate + SAN extension missing (either/or, based on the runtime environment) + type: Counter + stabilityLevel: ALPHA - name: cloudprovider_aws_api_request_duration_seconds help: Latency of AWS API calls type: Histogram @@ -3668,6 +3855,12 @@ - resource_group - source - subscription_id +- name: number_of_l4_ilbs + help: Number of L4 ILBs + type: Gauge + stabilityLevel: ALPHA + labels: + - feature - name: cloudprovider_gce_api_request_duration_seconds help: Latency of a GCE API call type: Histogram @@ -3726,12 +3919,6 @@ help: Counter of failed Token() requests to the alternate token source type: Counter stabilityLevel: ALPHA -- name: number_of_l4_ilbs - help: Number of L4 ILBs - type: Gauge - stabilityLevel: ALPHA - labels: - - feature - 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 3111818ed34..d0a1f007b13 100644 --- a/test/instrumentation/documentation/documentation.md +++ b/test/instrumentation/documentation/documentation.md @@ -7,7 +7,7 @@ description: >- ## Metrics (v1.26) - + 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. @@ -98,6 +98,41 @@ components using an HTTP scrape, and fetch the current metrics data in Prometheu
resource
+cronjob_controller_job_creation_skew_duration_seconds +STABLE +Histogram +Time between when a cronjob is scheduled to be run, and when the corresponding job is created + + + +job_controller_job_pods_finished_total +STABLE +Counter +The number of finished Pods that are fully tracked +
completion_mode
result
+ + +job_controller_job_sync_duration_seconds +STABLE +Histogram +The time it took to sync a job +
action
completion_mode
result
+ + +job_controller_job_syncs_total +STABLE +Counter +The number of job syncs +
action
completion_mode
result
+ + +job_controller_jobs_finished_total +STABLE +Counter +The number of finished jobs +
completion_mode
reason
result
+ + node_collector_evictions_total STABLE Counter @@ -115,7 +150,7 @@ components using an HTTP scrape, and fetch the current metrics data in Prometheu scheduler_pending_pods STABLE Gauge -Number of pending pods, by the queue type. 'active' means number of pods in activeQ; 'backoff' means number of pods in backoffQ; 'unschedulable' means number of pods in unschedulablePods. +Number of pending pods, by the queue type. 'active' means number of pods in activeQ; 'backoff' means number of pods in backoffQ; 'unschedulable' means number of pods in unschedulablePods that the scheduler attempted to schedule and failed; 'gated' is the number of unschedulable pods that the scheduler never attempted to schedule because they are gated.
queue
@@ -432,6 +467,13 @@ components using an HTTP scrape, and fetch the current metrics data in Prometheu
flow_schema
priority_level
+apiserver_flowcontrol_current_limit_seats +ALPHA +Gauge +current derived number of execution seats available to each priority level +
priority_level
+ + apiserver_flowcontrol_current_r ALPHA Gauge @@ -439,6 +481,41 @@ components using an HTTP scrape, and fetch the current metrics data in Prometheu
priority_level
+apiserver_flowcontrol_demand_seats +ALPHA +TimingRatioHistogram +Observations, at the end of every nanosecond, of (the number of seats each priority level could use) / (nominal number of seats for that level) +
priority_level
+ + +apiserver_flowcontrol_demand_seats_average +ALPHA +Gauge +Time-weighted average, over last adjustment period, of demand_seats +
priority_level
+ + +apiserver_flowcontrol_demand_seats_high_watermark +ALPHA +Gauge +High watermark, over last adjustment period, of demand_seats +
priority_level
+ + +apiserver_flowcontrol_demand_seats_smoothed +ALPHA +Gauge +Smoothed seat demands +
priority_level
+ + +apiserver_flowcontrol_demand_seats_stdev +ALPHA +Gauge +Time-weighted standard deviation, over last adjustment period, of demand_seats +
priority_level
+ + apiserver_flowcontrol_dispatch_r ALPHA Gauge @@ -467,6 +544,13 @@ components using an HTTP scrape, and fetch the current metrics data in Prometheu
priority_level
+apiserver_flowcontrol_lower_limit_seats +ALPHA +Gauge +Configured lower bound on number of execution seats available to each priority level +
priority_level
+ + apiserver_flowcontrol_next_discounted_s_bounds ALPHA Gauge @@ -481,6 +565,13 @@ components using an HTTP scrape, and fetch the current metrics data in Prometheu
bound
priority_level
+apiserver_flowcontrol_nominal_limit_seats +ALPHA +Gauge +Nominal number of execution seats configured for each priority level +
priority_level
+ + apiserver_flowcontrol_priority_level_request_utilization ALPHA TimingRatioHistogram @@ -551,6 +642,27 @@ components using an HTTP scrape, and fetch the current metrics data in Prometheu
execute
flow_schema
priority_level
+apiserver_flowcontrol_seat_fair_frac +ALPHA +Gauge +Fair fraction of server's concurrency to allocate to each priority level that can use it + + + +apiserver_flowcontrol_target_seats +ALPHA +Gauge +Seat allocation targets +
priority_level
+ + +apiserver_flowcontrol_upper_limit_seats +ALPHA +Gauge +Configured upper bound on number of execution seats available to each priority level +
priority_level
+ + apiserver_flowcontrol_watch_count_samples ALPHA Histogram @@ -614,13 +726,20 @@ components using an HTTP scrape, and fetch the current metrics data in Prometheu
source
status
-apiserver_request_slo_duration_seconds +apiserver_request_sli_duration_seconds ALPHA Histogram Response latency distribution (not counting webhook duration) in seconds for each verb, group, version, resource, subresource, scope and component.
component
group
resource
scope
subresource
verb
version
+apiserver_request_slo_duration_seconds +ALPHA +Histogram +Response latency distribution (not counting webhook duration) in seconds for each verb, group, version, resource, subresource, scope and component. +
component
group
resource
scope
subresource
verb
version
+ +1.27.0 apiserver_request_terminations_total ALPHA Counter @@ -985,13 +1104,6 @@ components using an HTTP scrape, and fetch the current metrics data in Prometheu
container
pod
namespace
-cronjob_controller_cronjob_job_creation_skew_duration_seconds -ALPHA -Histogram -Time between when a cronjob is scheduled to be run, and when the corresponding job is created - - - csi_operations_seconds ALPHA Histogram @@ -1188,34 +1300,6 @@ components using an HTTP scrape, and fetch the current metrics data in Prometheu -job_controller_job_finished_total -ALPHA -Counter -The number of finished job -
completion_mode
reason
result
- - -job_controller_job_pods_finished_total -ALPHA -Counter -The number of finished Pods that are fully tracked -
completion_mode
result
- - -job_controller_job_sync_duration_seconds -ALPHA -Histogram -The time it took to sync a job -
action
completion_mode
result
- - -job_controller_job_sync_total -ALPHA -Counter -The number of job syncs -
action
completion_mode
result
- - job_controller_pod_failures_handled_by_failure_policy_total ALPHA Counter @@ -1517,6 +1601,13 @@ components using an HTTP scrape, and fetch the current metrics data in Prometheu +kubelet_pod_start_sli_duration_seconds +ALPHA +Histogram +Duration in seconds to start a pod, excluding time to pull images and run init containers, measured from pod creation timestamp to when all its containers are reported as started and observed via watch + + + kubelet_pod_status_sync_duration_seconds ALPHA Histogram @@ -1944,6 +2035,20 @@ components using an HTTP scrape, and fetch the current metrics data in Prometheu
pod
namespace
+pod_gc_collector_force_delete_pod_errors_total +ALPHA +Counter +Number of errors encountered when forcefully deleting the pods since the Pod GC Controller started. + + + +pod_gc_collector_force_delete_pods_total +ALPHA +Counter +Number of pods that are being forcefully deleted since the Pod GC Controller started. + + + pod_memory_working_set_bytes ALPHA Custom diff --git a/test/instrumentation/error.go b/test/instrumentation/error.go index 4cee63be64e..b7e628358d7 100644 --- a/test/instrumentation/error.go +++ b/test/instrumentation/error.go @@ -26,7 +26,6 @@ const ( errNotDirectCall = "Opts for STABLE metric was not directly passed to new metric function" errPositionalArguments = "Positional arguments are not supported" errStabilityLevel = "StabilityLevel should be passed STABLE, ALPHA or removed" - errStableSummary = "Stable summary metric is not supported" errInvalidNewMetricCall = "Invalid new metric call, please ensure code compiles" errNonStringAttribute = "Non string attribute is not supported" errBadVariableAttribute = "Metric attribute was not correctly set. Please use only global consts in same file" @@ -40,6 +39,13 @@ const ( errLabels = "Labels were not set to list of strings" errImport = `Importing using "." is not supported` errExprNotIdent = "expr selector does not refer to type ast.Ident, is type %s" + + errorDecodingString = "can't decode string" + errorDecodingLabels = "can't decode labels" + errorDecodingConstLabels = "can't decode const labels" + errorDecodingStabilityLevel = "can't decode stability level" + errorFindingVariableForBuckets = "couldn't find variable for bucket" + errorFindingVariableForLabels = "couldn't find variable for labels" ) type decodeError struct { diff --git a/test/instrumentation/find_stable_metric.go b/test/instrumentation/find_stable_metric.go index a2b9f0af093..1e2590186ce 100644 --- a/test/instrumentation/find_stable_metric.go +++ b/test/instrumentation/find_stable_metric.go @@ -80,7 +80,6 @@ func (f *stableMetricFinder) Visit(node ast.Node) (w ast.Visitor) { return nil } } - } else { f.currentFunctionCall = opts }