The highest tagged major version is
Documentation
¶
Index ¶
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
func NewSplunkSpanSink ¶
func NewSplunkSpanSink(server string, token string, localHostname string, validateServerName string, log *logrus.Logger, ingestTimeout time.Duration, sendTimeout time.Duration, batchSize int, workers int, spanSampleRate int, maxConnLifetime time.Duration, connLifetimeJitter time.Duration) (sinks.SpanSink, error)
NewSplunkSpanSink constructs a new splunk span sink from the server name and token provided, using the local hostname configured for veneur. An optional argument, validateServerName is used (if non-empty) to instruct go to validate a different hostname than the one on the server URL. The spanSampleRate is an integer. For any given trace ID, the probability that all spans in the trace will be chosen for the sample is 1/spanSampleRate. Sampling is performed on the trace ID, so either all spans within a given trace will be chosen, or none will.
Types ¶
type Event ¶
type Event struct {
Host *string `json:"host,omitempty"`
Index *string `json:"index,omitempty"`
Source *string `json:"source,omitempty"`
SourceType *string `json:"sourcetype,omitempty"`
Time *string `json:"time,omitempty"`
Event interface{} `json:"event"`
}
func (*Event) SetSourceType ¶
type Response ¶
type Response struct {
Text string `json:"text,omitempty"`
Code int `json:"code"`
InvalidEventNumber *int `json:"invalid-event-number,omitempty"`
}
Response represents the JSON-parseable response from a splunk HEC server.
type SerializedSSF ¶
type SerializedSSF struct {
TraceId string `json:"trace_id"`
Id string `json:"id"`
ParentId string `json:"parent_id"`
StartTimestamp float64 `json:"start_timestamp"`
EndTimestamp float64 `json:"end_timestamp"`
Duration int64 `json:"duration_ns"`
Error bool `json:"error"`
Service string `json:"service"`
Tags map[string]string `json:"tags"`
Indicator bool `json:"indicator"`
Name string `json:"name"`
Partial *bool `json:"partial,omitempty"`
}
SerializedSSF holds a set of fields in a format that Splunk can handle (it can't handle int64s, and we don't want to round our traceID to the thousands place). This is mildly redundant, but oh well.
type TestableSplunkSpanSink ¶
type TestableSplunkSpanSink interface {
sinks.SpanSink
// Stop shuts down the sink's submission workers by finishing
// each worker's last submission HTTP request.
Stop()
// Sync instructs all submission workers to finish submitting
// their current request and start a new one. It returns when
// the last worker's submission is done.
Sync()
}
TestableSplunkSpanSink provides methods that are useful for testing a splunk span sink.
Source Files