Documentation
¶
Index ¶
Constants ¶
View Source
const ( TrueValue = 1.0 FalseValue = 0.0 )
View Source
const ( // UntaintTimeoutReasonReadiness signals that a pod existed on the node but // never became Ready within --untaintControllerTimeout. UntaintTimeoutReasonReadiness = "readiness" // UntaintTimeoutReasonScheduling signals that no agent pod was scheduled on // the node within --untaintControllerSchedulingTimeout. UntaintTimeoutReasonScheduling = "scheduling" // UntaintTimeoutPolicyRemove untaints the node despite the agent not being ready. UntaintTimeoutPolicyRemove = "remove" // UntaintTimeoutPolicyKeep leaves the taint in place but emits observability signals. UntaintTimeoutPolicyKeep = "keep" )
Label values for TaintTimeoutsTotal.
Variables ¶
View Source
var ( // datadogagentprofile enabled DAPEnabled = prometheus.NewGauge( prometheus.GaugeOpts{ Subsystem: datadogAgentProfileSubsystem, Name: "enabled", Help: "1 if DatadogAgentProfiles are enabled. 0 if DatadogAgentProfiles are disabled", }, ) // datadogagentprofile valid DAPValid = prometheus.NewGaugeVec( prometheus.GaugeOpts{ Subsystem: datadogAgentProfileSubsystem, Name: "valid", Help: "1 if the DatadogAgentProfile is valid. 0 if the DatadogAgentProfile is invalid", }, []string{ datadogAgentProfileLabelKey, }, ) )
View Source
var ( // TaintRemovalsTotal is the total number of taints removed from nodes. TaintRemovalsTotal = prometheus.NewCounter( prometheus.CounterOpts{ Subsystem: untaintSubsystem, Name: "taint_removals_total", Help: "Total number of taints removed from nodes", }, ) // TaintRemovalLatency is the time between agent pod becoming Ready and taint removal. TaintRemovalLatency = prometheus.NewHistogram( prometheus.HistogramOpts{ Subsystem: untaintSubsystem, Name: "taint_removal_latency_seconds", Help: "Time between agent pod becoming Ready and taint removal from the node", Buckets: prometheus.DefBuckets, }, ) // TaintTimeoutsTotal counts timeout decisions broken down by reason and policy. TaintTimeoutsTotal = prometheus.NewCounterVec( prometheus.CounterOpts{ Subsystem: untaintSubsystem, Name: "taint_timeouts_total", Help: "Total number of untaint-controller timeout decisions, by reason and policy", }, []string{"reason", "policy"}, ) // TaintRemovalErrorsTotal counts hard errors encountered while attempting to // remove the taint (apiserver Patch failures, JSON marshal failures, …). // Benign optimistic-concurrency races (IsConflict/IsInvalid) are NOT counted // here — they're handled by requeueing. Inspect the operator's ERROR-level // logs for the specific failure cause. TaintRemovalErrorsTotal = prometheus.NewCounter( prometheus.CounterOpts{ Subsystem: untaintSubsystem, Name: "taint_removal_errors_total", Help: "Total number of errors encountered while attempting to remove the agent-not-ready taint from a node", }, ) )
View Source
var ( // introspection enabled IntrospectionEnabled = prometheus.NewGauge( prometheus.GaugeOpts{ Subsystem: datadogAgentSubsystem, Name: "introspection_enabled", Help: "1 if introspection is enabled. 0 if introspection is disabled", }, ) )
View Source
var ( // maximum goroutines MaxGoroutines = prometheus.NewGauge( prometheus.GaugeOpts{ Name: "max_goroutines", Help: "reports the maximum number of goroutines set in the datadog operator", }, ) )
Functions ¶
func CleanupMetricsByProfile ¶
CleanupMetricsByProfile deletes profile-specific prometheus metrics given a profile
Types ¶
This section is empty.
Click to show internal directories.
Click to hide internal directories.