VictoriaMetrics/lib/streamaggr/rate.go
Aliaksandr Valialkin cbc637d1dd
app/vmagent/remotewrite: follow-up for f153f54d11
- Move the remaining code responsible for stream aggregation initialization from remotewrite.go to streamaggr.go .
  This improves code maintainability a bit.

- Properly shut down streamaggr.Aggregators initialized inside remotewrite.CheckStreamAggrConfigs().
  This prevents from potential resource leaks.

- Use separate functions for initializing and reloading of global stream aggregation and per-remoteWrite.url stream aggregation.
  This makes the code easier to read and maintain. This also fixes INFO and ERROR logs emitted by these functions.

- Add an ability to specify `name` option in every stream aggregation config. This option is used as `name` label
  in metrics exposed by stream aggregation at /metrics page. This simplifies investigation of the exposed metrics.

- Add `path` label additionally to `name`, `url` and `position` labels at metrics exposed by streaming aggregation.
  This label should simplify investigation of the exposed metrics.

- Remove `match` and `group` labels from metrics exposed by streaming aggregation, since they have little practical applicability:
  it is hard to use these labels in query filters and aggregation functions.

- Rename the metric `vm_streamaggr_flushed_samples_total` to less misleading `vm_streamaggr_output_samples_total` .
  This metric shows the number of samples generated by the corresponding streaming aggregation rule.
  This metric has been added in the commit 861852f262 .
  See https://github.com/VictoriaMetrics/VictoriaMetrics/pull/6462

- Remove the metric `vm_streamaggr_stale_samples_total`, since it is unclear how it can be used in practice.
  This metric has been added in the commit 861852f262 .
  See https://github.com/VictoriaMetrics/VictoriaMetrics/pull/6462

- Remove Alias and aggrID fields from streamaggr.Options struct, since these fields aren't related to optional params,
  which could modify the behaviour of the constructed streaming aggregator.
  Convert the Alias field to regular argument passed to LoadFromFile() function, since this argument is mandatory.

- Pass Options arg to LoadFromFile() function by reference, since this structure is quite big.
  This also allows passing nil instead of Options when default options are enough.

- Add `name`, `path`, `url` and `position` labels to `vm_streamaggr_dedup_state_size_bytes` and `vm_streamaggr_dedup_state_items_count` metrics,
  so they have consistent set of labels comparing to the rest of streaming aggregation metrics.

- Convert aggregator.aggrStates field type from `map[string]aggrState` to `[]aggrOutput`, where `aggrOutput` contains the corresponding
  `aggrState` plus all the related metrics (currently only `vm_streamaggr_output_samples_total` metric is exposed with the corresponding
  `output` label per each configured output function). This simplifies and speeds up the code responsible for updating per-output
  metrics. This is a follow-up for the commit 2eb1bc4f81 .
  See https://github.com/VictoriaMetrics/VictoriaMetrics/pull/6604

- Added missing urls to docs ( https://docs.victoriametrics.com/stream-aggregation/ ) in error messages. These urls help users
  figuring out why VictoriaMetrics or vmagent generates the corresponding error messages. The urls were removed for unknown reason
  in the commit 2eb1bc4f81 .

- Fix incorrect update for `vm_streamaggr_output_samples_total` metric in flushCtx.appendSeriesWithExtraLabel() function.
  While at it, reduce memory usage by limiting the maximum number of samples per flush to 10K.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/5467
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/pull/6268
2024-07-15 20:25:36 +02:00

186 lines
4.1 KiB
Go

package streamaggr
import (
"sync"
"time"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/fasttime"
)
// rateAggrState calculates output=rate_avg and rate_sum, e.g. the average per-second increase rate for counter metrics.
type rateAggrState struct {
m sync.Map
// isAvg is set to true if rate_avg() must be calculated instead of rate_sum().
isAvg bool
// Time series state is dropped if no new samples are received during stalenessSecs.
stalenessSecs uint64
}
type rateStateValue struct {
mu sync.Mutex
lastValues map[string]rateLastValueState
deleteDeadline uint64
deleted bool
}
type rateLastValueState struct {
value float64
timestamp int64
deleteDeadline uint64
// increase stores cumulative increase for the current time series on the current aggregation interval
increase float64
// prevTimestamp is the timestamp of the last registered sample in the previous aggregation interval
prevTimestamp int64
}
func newRateAggrState(stalenessInterval time.Duration, isAvg bool) *rateAggrState {
stalenessSecs := roundDurationToSecs(stalenessInterval)
return &rateAggrState{
isAvg: isAvg,
stalenessSecs: stalenessSecs,
}
}
func (as *rateAggrState) pushSamples(samples []pushSample) {
currentTime := fasttime.UnixTimestamp()
deleteDeadline := currentTime + as.stalenessSecs
for i := range samples {
s := &samples[i]
inputKey, outputKey := getInputOutputKey(s.key)
again:
v, ok := as.m.Load(outputKey)
if !ok {
// The entry is missing in the map. Try creating it.
v = &rateStateValue{
lastValues: make(map[string]rateLastValueState),
}
outputKey = bytesutil.InternString(outputKey)
vNew, loaded := as.m.LoadOrStore(outputKey, v)
if loaded {
// Use the entry created by a concurrent goroutine.
v = vNew
}
}
sv := v.(*rateStateValue)
sv.mu.Lock()
deleted := sv.deleted
if !deleted {
lv, ok := sv.lastValues[inputKey]
if ok {
if s.timestamp < lv.timestamp {
// Skip out of order sample
sv.mu.Unlock()
continue
}
if s.value >= lv.value {
lv.increase += s.value - lv.value
} else {
// counter reset
lv.increase += s.value
}
} else {
lv.prevTimestamp = s.timestamp
}
lv.value = s.value
lv.timestamp = s.timestamp
lv.deleteDeadline = deleteDeadline
inputKey = bytesutil.InternString(inputKey)
sv.lastValues[inputKey] = lv
sv.deleteDeadline = deleteDeadline
}
sv.mu.Unlock()
if deleted {
// The entry has been deleted by the concurrent call to flushState
// Try obtaining and updating the entry again.
goto again
}
}
}
func (as *rateAggrState) flushState(ctx *flushCtx, resetState bool) {
currentTime := fasttime.UnixTimestamp()
currentTimeMsec := int64(currentTime) * 1000
suffix := as.getSuffix()
as.removeOldEntries(currentTime)
m := &as.m
m.Range(func(k, v any) bool {
sv := v.(*rateStateValue)
sv.mu.Lock()
lvs := sv.lastValues
sumRate := 0.0
countSeries := 0
for k1, lv := range lvs {
d := float64(lv.timestamp-lv.prevTimestamp) / 1000
if d > 0 {
sumRate += lv.increase / d
countSeries++
}
if resetState {
lv.prevTimestamp = lv.timestamp
lv.increase = 0
lvs[k1] = lv
}
}
deleted := sv.deleted
sv.mu.Unlock()
if countSeries == 0 || deleted {
// Nothing to update
return true
}
result := sumRate
if as.isAvg {
result /= float64(countSeries)
}
key := k.(string)
ctx.appendSeries(key, suffix, currentTimeMsec, result)
return true
})
}
func (as *rateAggrState) getSuffix() string {
if as.isAvg {
return "rate_avg"
}
return "rate_sum"
}
func (as *rateAggrState) removeOldEntries(currentTime uint64) {
m := &as.m
m.Range(func(k, v any) bool {
sv := v.(*rateStateValue)
sv.mu.Lock()
if currentTime > sv.deleteDeadline {
// Mark the current entry as deleted
sv.deleted = true
sv.mu.Unlock()
m.Delete(k)
return true
}
// Delete outdated entries in sv.lastValues
lvs := sv.lastValues
for k1, lv := range lvs {
if currentTime > lv.deleteDeadline {
delete(lvs, k1)
}
}
sv.mu.Unlock()
return true
})
}