|
| 1 | +package health |
| 2 | + |
| 3 | +import ( |
| 4 | + "time" |
| 5 | +) |
| 6 | + |
| 7 | +type aggregator struct { |
| 8 | + // How long is each aggregation interval. Eg, 1 minute |
| 9 | + intervalDuration time.Duration |
| 10 | + |
| 11 | + // Retain controls how many metrics interval we keep. Eg, 5 minutes |
| 12 | + retain time.Duration |
| 13 | + |
| 14 | + // maxIntervals is the maximum length of intervals. |
| 15 | + // It is retain / interval. |
| 16 | + maxIntervals int |
| 17 | + |
| 18 | + // intervals is a slice of the retained intervals |
| 19 | + intervalAggregations []*IntervalAggregation |
| 20 | +} |
| 21 | + |
| 22 | +func startAggregator(intervalDuration time.Duration, retain time.Duration, sink *JsonPollingSink) { |
| 23 | + cmdChan := sink.cmdChan |
| 24 | + doneChan := sink.doneChan |
| 25 | + intervalsChanChan := sink.intervalsChanChan |
| 26 | + ticker := time.Tick(1 * time.Second) |
| 27 | + |
| 28 | + agg := newAggregator(intervalDuration, retain) |
| 29 | + |
| 30 | + for { |
| 31 | + select { |
| 32 | + case <-doneChan: |
| 33 | + break |
| 34 | + case cmd := <-cmdChan: |
| 35 | + if cmd.Kind == cmdKindEvent { |
| 36 | + agg.EmitEvent(cmd.Job, cmd.Event) |
| 37 | + } else if cmd.Kind == cmdKindEventErr { |
| 38 | + agg.EmitEventErr(cmd.Job, cmd.Event, cmd.Err) |
| 39 | + } else if cmd.Kind == cmdKindTiming { |
| 40 | + agg.EmitTiming(cmd.Job, cmd.Event, cmd.Nanos) |
| 41 | + } else if cmd.Kind == cmdKindComplete { |
| 42 | + agg.EmitComplete(cmd.Job, cmd.Status, cmd.Nanos) |
| 43 | + } |
| 44 | + case <-ticker: |
| 45 | + agg.getIntervalAggregation() // this has the side effect of sliding the interval window if necessary. |
| 46 | + case intervalsChan := <-intervalsChanChan: |
| 47 | + intervalsChan <- agg.memorySafeIntervals() |
| 48 | + } |
| 49 | + } |
| 50 | +} |
| 51 | + |
| 52 | +func newAggregator(intervalDuration time.Duration, retain time.Duration) *aggregator { |
| 53 | + maxIntervals := int(retain / intervalDuration) |
| 54 | + return &aggregator{ |
| 55 | + intervalDuration: intervalDuration, |
| 56 | + retain: retain, |
| 57 | + maxIntervals: maxIntervals, |
| 58 | + intervalAggregations: make([]*IntervalAggregation, 0, maxIntervals), |
| 59 | + } |
| 60 | +} |
| 61 | + |
| 62 | +func (a *aggregator) memorySafeIntervals() []*IntervalAggregation { |
| 63 | + ret := make([]*IntervalAggregation, 0, len(a.intervalAggregations)) |
| 64 | + curAgg := a.getIntervalAggregation() |
| 65 | + |
| 66 | + for _, intAgg := range a.intervalAggregations { |
| 67 | + if intAgg == curAgg { |
| 68 | + ret = append(ret, intAgg.Clone()) |
| 69 | + } else { |
| 70 | + ret = append(ret, intAgg) |
| 71 | + } |
| 72 | + } |
| 73 | + |
| 74 | + return ret |
| 75 | +} |
| 76 | + |
| 77 | +func (a *aggregator) EmitEvent(job string, event string) { |
| 78 | + intAgg := a.getIntervalAggregation() |
| 79 | + intAgg.Events[event] = intAgg.Events[event] + 1 |
| 80 | + jobAgg := intAgg.getJobAggregation(job) |
| 81 | + jobAgg.Events[event] = jobAgg.Events[event] + 1 |
| 82 | + intAgg.SerialNumber++ |
| 83 | +} |
| 84 | + |
| 85 | +func (a *aggregator) EmitEventErr(job string, event string, inputErr error) { |
| 86 | + intAgg := a.getIntervalAggregation() |
| 87 | + errc := intAgg.getCounterErrs(event) |
| 88 | + errc.incrementAndAddError(inputErr) |
| 89 | + jobAgg := intAgg.getJobAggregation(job) |
| 90 | + jerrc := jobAgg.getCounterErrs(event) |
| 91 | + jerrc.incrementAndAddError(inputErr) |
| 92 | + intAgg.SerialNumber++ |
| 93 | +} |
| 94 | + |
| 95 | +func (a *aggregator) EmitTiming(job string, event string, nanos int64) { |
| 96 | + intAgg := a.getIntervalAggregation() |
| 97 | + t := intAgg.getTimers(event) |
| 98 | + t.ingest(nanos) |
| 99 | + jobAgg := intAgg.getJobAggregation(job) |
| 100 | + jt := jobAgg.getTimers(event) |
| 101 | + jt.ingest(nanos) |
| 102 | + intAgg.SerialNumber++ |
| 103 | +} |
| 104 | + |
| 105 | +func (a *aggregator) EmitComplete(job string, status CompletionStatus, nanos int64) { |
| 106 | + intAgg := a.getIntervalAggregation() |
| 107 | + jobAgg := intAgg.getJobAggregation(job) |
| 108 | + jobAgg.ingest(status, nanos) |
| 109 | + intAgg.SerialNumber++ |
| 110 | +} |
| 111 | + |
| 112 | +func (a *aggregator) getIntervalAggregation() *IntervalAggregation { |
| 113 | + intervalStart := now().Truncate(a.intervalDuration) |
| 114 | + |
| 115 | + n := len(a.intervalAggregations) |
| 116 | + if n > 0 && a.intervalAggregations[n-1].IntervalStart == intervalStart { |
| 117 | + return a.intervalAggregations[n-1] |
| 118 | + } |
| 119 | + |
| 120 | + return a.createIntervalAggregation(intervalStart) |
| 121 | +} |
| 122 | + |
| 123 | +func (a *aggregator) createIntervalAggregation(interval time.Time) *IntervalAggregation { |
| 124 | + // Make new interval: |
| 125 | + current := NewIntervalAggregation(interval) |
| 126 | + |
| 127 | + // If we've reached our max intervals, and we're going to shift everything down, then set the last one |
| 128 | + n := len(a.intervalAggregations) |
| 129 | + if n == a.maxIntervals { |
| 130 | + for i := 1; i < n; i++ { |
| 131 | + a.intervalAggregations[i-1] = a.intervalAggregations[i] |
| 132 | + } |
| 133 | + a.intervalAggregations[n-1] = current |
| 134 | + } else { |
| 135 | + a.intervalAggregations = append(a.intervalAggregations, current) |
| 136 | + } |
| 137 | + |
| 138 | + return current |
| 139 | +} |
| 140 | + |
| 141 | +var nowMock time.Time |
| 142 | + |
| 143 | +func now() time.Time { |
| 144 | + if nowMock.IsZero() { |
| 145 | + return time.Now() |
| 146 | + } |
| 147 | + return nowMock |
| 148 | +} |
| 149 | + |
| 150 | +func setNowMock(t string) { |
| 151 | + var err error |
| 152 | + nowMock, err = time.Parse(time.RFC3339, t) |
| 153 | + if err != nil { |
| 154 | + panic(err) |
| 155 | + } |
| 156 | +} |
| 157 | + |
| 158 | +func resetNowMock() { |
| 159 | + nowMock = time.Time{} |
| 160 | +} |
0 commit comments