Documentation
¶
Index ¶
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
func PrometheusEndpoint ¶
func PrometheusEndpoint(ctx context.Context, cfg *PrometheusConfig, promMgr *connector.PrometheusManager) (pipe.FinalFunc[[]*ebpf.Record], error)
Types ¶
type Expirer ¶
type Expirer struct {
// contains filtered or unexported fields
}
Expirer drops metrics from labels that haven't been updated during a given timeout TODO: generify and move to a common section for using it also in AppO11y, supporting more prometheus metrics
func NewExpirer ¶
func NewExpirer(wrapped *prometheus.CounterVec, expireTime time.Duration) *Expirer
NewExpirer creates a metric that wraps a given CounterVec. Its labeled instances are dropped if they haven't been updated during the last timeout period
func (*Expirer) Collect ¶
func (ex *Expirer) Collect(metrics chan<- prometheus.Metric)
Collect wraps prometheus.Collector Wrap method
func (*Expirer) Describe ¶
func (ex *Expirer) Describe(descs chan<- *prometheus.Desc)
Describe wraps prometheus.Collector Describe method
func (*Expirer) UpdateTime ¶
func (ex *Expirer) UpdateTime()
UpdateTime updates the last access time to be annotated to any new or existing metric. It is a required operation before processing a given batch of metrics (invoking the WithLabelValues).
func (*Expirer) WithLabelValues ¶
func (ex *Expirer) WithLabelValues(lbls ...string) prometheus.Counter
WithLabelValues returns the Counter for the given slice of label values (same order as the variable labels in Desc). If that combination of label values is accessed for the first time, a new Counter is created. If not, a cached copy is returned and the "last access" cache time is updated.
type PrometheusConfig ¶
type PrometheusConfig struct {
Config *prom.PrometheusConfig
AllowedAttributes []string
}
PrometheusConfig for network metrics just wraps the global prom.PrometheusConfig as provided by the user
Source Files