metricsutil

Documentation

Index

• Constants
• func FormatFromRequest(req *logical.Request) string
• func TTLBucket(ttl time.Duration) string
• type ClusterMetricSink
  • func BlackholeSink() *ClusterMetricSink
  • func NewClusterMetricSink(clusterName string, sink metrics.MetricSink) *ClusterMetricSink
  • func SetupTelemetry(opts *SetupTelemetryOpts) (*metrics.InmemSink, *ClusterMetricSink, bool, error)
  • func (m *ClusterMetricSink) AddDurationWithLabels(key []string, d time.Duration, labels []Label)
  • func (m *ClusterMetricSink) AddSampleWithLabels(key []string, val float32, labels []Label)
  • func (m *ClusterMetricSink) IncrCounterWithLabels(key []string, val float32, labels []Label)
  • func (m *ClusterMetricSink) MeasureSinceWithLabels(key []string, start time.Time, labels []Label)
  • func (m *ClusterMetricSink) NewGaugeCollectionProcess(key []string, id []Label, collector GaugeCollector, logger log.Logger) (*GaugeCollectionProcess, error)
  • func (m *ClusterMetricSink) SetDefaultClusterName(clusterName string)
  • func (m *ClusterMetricSink) SetGaugeWithLabels(key []string, val float32, labels []Label)
• type GaugeCollectionProcess
  • func (p *GaugeCollectionProcess) Run()
  • func (p *GaugeCollectionProcess) Stop()
• type GaugeCollector
• type GaugeLabelValues
• type Label
• type MetricsHelper
  • func NewMetricsHelper(inMem *metrics.InmemSink, enablePrometheus bool) *MetricsHelper
  • func (m *MetricsHelper) GenericResponse() *logical.Response
  • func (m *MetricsHelper) PrometheusResponse() *logical.Response
  • func (m *MetricsHelper) ResponseForFormat(format string) *logical.Response
• type SetupTelemetryOpts
• type Telemetry
  • func (t *Telemetry) GoString() string

Constants

const (
	OpenMetricsMIMEType = "application/openmetrics-text"

	PrometheusSchemaMIMEType = "prometheus/telemetry"

	// ErrorContentType is the content type returned by an error response.
	ErrorContentType = "text/plain"
)

const (
	PrometheusDefaultRetentionTime = 24 * time.Hour
	UsageGaugeDefaultPeriod        = 10 * time.Minute
	MaximumGaugeCardinalityDefault = 500
)

const OverflowBucket = "+Inf"

const (
	PrometheusMetricFormat = "prometheus"
)

Functions

func FormatFromRequest

func FormatFromRequest(req *logical.Request) string

func TTLBucket

func TTLBucket(ttl time.Duration) string

TTLBucket computes the label to apply for a token TTL.

Types

type ClusterMetricSink

type ClusterMetricSink struct {
	// ClusterName is either the cluster ID, or a name provided
	// in the telemetry configuration stanza.
	//
	// Because it may be set after the Core is initialized, we need
	// to protect against concurrent access.
	ClusterName atomic.Value

	MaxGaugeCardinality int
	GaugeInterval       time.Duration

	// Sink is the go-metrics instance to send to.
	Sink metrics.MetricSink
}

ClusterMetricSink serves as a shim around go-metrics and inserts a "cluster" label.

It also provides a mechanism to limit the cardinality of the labels on a gauge (at each reporting interval, which isn't sufficient if there is variability in which labels are the top N) and a backoff mechanism for gauge computation.

func BlackholeSink

func BlackholeSink() *ClusterMetricSink

BlackholeSink is a default suitable for use in unit tests.

func NewClusterMetricSink

func NewClusterMetricSink(clusterName string, sink metrics.MetricSink) *ClusterMetricSink

func SetupTelemetry

func SetupTelemetry(opts *SetupTelemetryOpts) (*metrics.InmemSink, *ClusterMetricSink, bool, error)

SetupTelemetry is used to setup the telemetry sub-systems and returns the in-memory sink to be used in http configuration

func (*ClusterMetricSink) AddDurationWithLabels

func (m *ClusterMetricSink) AddDurationWithLabels(key []string, d time.Duration, labels []Label)

func (*ClusterMetricSink) AddSampleWithLabels

func (m *ClusterMetricSink) AddSampleWithLabels(key []string, val float32, labels []Label)

func (*ClusterMetricSink) IncrCounterWithLabels

func (m *ClusterMetricSink) IncrCounterWithLabels(key []string, val float32, labels []Label)

func (*ClusterMetricSink) MeasureSinceWithLabels

func (m *ClusterMetricSink) MeasureSinceWithLabels(key []string, start time.Time, labels []Label)

func (*ClusterMetricSink) NewGaugeCollectionProcess

func (m *ClusterMetricSink) NewGaugeCollectionProcess(
	key []string,
	id []Label,
	collector GaugeCollector,
	logger log.Logger,
) (*GaugeCollectionProcess, error)

NewGaugeCollectionProcess creates a new collection process for the callback function given as an argument, and starts it running. A label should be provided for metrics *about* this collection process.

The Run() method must be called to start the process.

func (*ClusterMetricSink) SetDefaultClusterName

func (m *ClusterMetricSink) SetDefaultClusterName(clusterName string)

SetDefaultClusterName changes the cluster name from its default value, if it has not previously been configured.

func (*ClusterMetricSink) SetGaugeWithLabels

func (m *ClusterMetricSink) SetGaugeWithLabels(key []string, val float32, labels []Label)

type GaugeCollectionProcess

type GaugeCollectionProcess struct {
	// contains filtered or unexported fields
}

A GaugeCollectionProcess is responsible for one particular gauge metric. It handles a delay on initial startup; limiting the cardinality; and exponential backoff on the requested interval.

func (*GaugeCollectionProcess) Run

func (p *GaugeCollectionProcess) Run()

Run should be called as a goroutine.

func (*GaugeCollectionProcess) Stop

func (p *GaugeCollectionProcess) Stop()

Stop the collection process

type GaugeCollector

type GaugeCollector = func(context.Context) ([]GaugeLabelValues, error)

GaugeCollector is a callback function that returns an unfiltered set of label-value pairs. It may be cancelled if it takes too long.

type GaugeLabelValues

type GaugeLabelValues struct {
	Labels []Label
	Value  float32
}

GaugeLabelValues is one gauge in a set sharing a single key, that are measured in a batch.

type Label

type Label = metrics.Label

Convenience alias

type MetricsHelper

type MetricsHelper struct {
	PrometheusEnabled bool
	// contains filtered or unexported fields
}

func NewMetricsHelper

func NewMetricsHelper(inMem *metrics.InmemSink, enablePrometheus bool) *MetricsHelper

func (*MetricsHelper) GenericResponse

func (m *MetricsHelper) GenericResponse() *logical.Response

func (*MetricsHelper) PrometheusResponse

func (m *MetricsHelper) PrometheusResponse() *logical.Response

func (*MetricsHelper) ResponseForFormat

func (m *MetricsHelper) ResponseForFormat(format string) *logical.Response

type SetupTelemetryOpts

type SetupTelemetryOpts struct {
	Config      *Telemetry
	Ui          cli.Ui
	ServiceName string
	DisplayName string
	UserAgent   string
	ClusterName string
}

type Telemetry

type Telemetry struct {
	StatsiteAddr string `hcl:"statsite_address"`
	StatsdAddr   string `hcl:"statsd_address"`

	DisableHostname     bool   `hcl:"disable_hostname"`
	EnableHostnameLabel bool   `hcl:"enable_hostname_label"`
	MetricsPrefix       string `hcl:"metrics_prefix"`
	UsageGaugePeriod    time.Duration
	UsageGaugePeriodRaw interface{} `hcl:"usage_gauge_period"`

	MaximumGaugeCardinality int `hcl:"maximum_gauge_cardinality"`

	// CirconusAPIToken is a valid API Token used to create/manage check. If provided,
	// metric management is enabled.
	// Default: none
	CirconusAPIToken string `hcl:"circonus_api_token"`
	// CirconusAPIApp is an app name associated with API token.
	// Default: "consul"
	CirconusAPIApp string `hcl:"circonus_api_app"`
	// CirconusAPIURL is the base URL to use for contacting the Circonus API.
	// Default: "https://api.circonus.com/v2"
	CirconusAPIURL string `hcl:"circonus_api_url"`
	// CirconusSubmissionInterval is the interval at which metrics are submitted to Circonus.
	// Default: 10s
	CirconusSubmissionInterval string `hcl:"circonus_submission_interval"`
	// CirconusCheckSubmissionURL is the check.config.submission_url field from a
	// previously created HTTPTRAP check.
	// Default: none
	CirconusCheckSubmissionURL string `hcl:"circonus_submission_url"`
	// CirconusCheckID is the check id (not check bundle id) from a previously created
	// HTTPTRAP check. The numeric portion of the check._cid field.
	// Default: none
	CirconusCheckID string `hcl:"circonus_check_id"`
	// CirconusCheckForceMetricActivation will force enabling metrics, as they are encountered,
	// if the metric already exists and is NOT active. If check management is enabled, the default
	// behavior is to add new metrics as they are encountered. If the metric already exists in the
	// check, it will *NOT* be activated. This setting overrides that behavior.
	// Default: "false"
	CirconusCheckForceMetricActivation string `hcl:"circonus_check_force_metric_activation"`
	// CirconusCheckInstanceID serves to uniquely identify the metrics coming from this "instance".
	// It can be used to maintain metric continuity with transient or ephemeral instances as
	// they move around within an infrastructure.
	// Default: hostname:app
	CirconusCheckInstanceID string `hcl:"circonus_check_instance_id"`
	// CirconusCheckSearchTag is a special tag which, when coupled with the instance id, helps to
	// narrow down the search results when neither a Submission URL or Check ID is provided.
	// Default: service:app (e.g. service:consul)
	CirconusCheckSearchTag string `hcl:"circonus_check_search_tag"`
	// CirconusCheckTags is a comma separated list of tags to apply to the check. Note that
	// the value of CirconusCheckSearchTag will always be added to the check.
	// Default: none
	CirconusCheckTags string `hcl:"circonus_check_tags"`
	// CirconusCheckDisplayName is the name for the check which will be displayed in the Circonus UI.
	// Default: value of CirconusCheckInstanceID
	CirconusCheckDisplayName string `hcl:"circonus_check_display_name"`
	// CirconusBrokerID is an explicit broker to use when creating a new check. The numeric portion
	// of broker._cid. If metric management is enabled and neither a Submission URL nor Check ID
	// is provided, an attempt will be made to search for an existing check using Instance ID and
	// Search Tag. If one is not found, a new HTTPTRAP check will be created.
	// Default: use Select Tag if provided, otherwise, a random Enterprise Broker associated
	// with the specified API token or the default Circonus Broker.
	// Default: none
	CirconusBrokerID string `hcl:"circonus_broker_id"`
	// CirconusBrokerSelectTag is a special tag which will be used to select a broker when
	// a Broker ID is not provided. The best use of this is to as a hint for which broker
	// should be used based on *where* this particular instance is running.
	// (e.g. a specific geo location or datacenter, dc:sfo)
	// Default: none
	CirconusBrokerSelectTag string `hcl:"circonus_broker_select_tag"`

	// Dogstats:
	// DogStatsdAddr is the address of a dogstatsd instance. If provided,
	// metrics will be sent to that instance
	DogStatsDAddr string `hcl:"dogstatsd_addr"`

	// DogStatsdTags are the global tags that should be sent with each packet to dogstatsd
	// It is a list of strings, where each string looks like "my_tag_name:my_tag_value"
	DogStatsDTags []string `hcl:"dogstatsd_tags"`

	// Prometheus:
	// PrometheusRetentionTime is the retention time for prometheus metrics if greater than 0.
	// Default: 24h
	PrometheusRetentionTime    time.Duration `hcl:"-"`
	PrometheusRetentionTimeRaw interface{}   `hcl:"prometheus_retention_time"`

	// Stackdriver:
	// StackdriverProjectID is the project to publish stackdriver metrics to.
	StackdriverProjectID string `hcl:"stackdriver_project_id"`
	// StackdriverLocation is the GCP or AWS region of the monitored resource.
	StackdriverLocation string `hcl:"stackdriver_location"`
	// StackdriverNamespace is the namespace identifier, such as a cluster name.
	StackdriverNamespace string `hcl:"stackdriver_namespace"`
	// StackdriverDebugLogs will write additional stackdriver related debug logs to stderr.
	StackdriverDebugLogs bool `hcl:"stackdriver_debug_logs"`
}

Telemetry is the telemetry configuration for the server

func (*Telemetry) GoString

func (t *Telemetry) GoString() string