app/vmselect/promql: do not extend too short lookbehind window for rate() function if it is set explicitly

Previously too short lookbehind window d for rate(m[d]) could be automatically extended
if it didn't cover at least two raw samples. This was needed in order to guarantee
non-empty results from rate(m[d]) on short time ranges.

Now the lookbehind window isn't extended if it is set explicitly,
since it is expected that the user knows what he is doing.

The lookbehind window continues to be extended when needed if it isn't set explicitly.
For example, in the case of rate(m).

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3483
This commit is contained in:
Aliaksandr Valialkin 2022-12-20 00:12:54 -08:00
parent eabb8762ee
commit 9fa3f1dc57
No known key found for this signature in database
GPG key ID: A72BEC6CD3D0DED1
3 changed files with 25 additions and 13 deletions

View file

@ -935,7 +935,7 @@ func TestExecSuccess(t *testing.T) {
))`
r := netstorage.Result{
MetricName: metricNameExpected,
Values: []float64{nan, nan, nan, 1, nan, nan},
Values: []float64{nan, nan, 1, 1, nan, nan},
Timestamps: timestampsExpected,
}
resultExpected := []netstorage.Result{r}
@ -6692,7 +6692,7 @@ func TestExecSuccess(t *testing.T) {
q := `rate((2000-time())[100s])`
r := netstorage.Result{
MetricName: metricNameExpected,
Values: []float64{5.5, 4.5, 3.5, 2.5, 1.5, 0.5},
Values: []float64{5, 4, 3, 2, 1, 0},
Timestamps: timestampsExpected,
}
resultExpected := []netstorage.Result{r}
@ -6703,7 +6703,7 @@ func TestExecSuccess(t *testing.T) {
q := `rate((2000-time())[100s:])`
r := netstorage.Result{
MetricName: metricNameExpected,
Values: []float64{5.5, 4.5, 3.5, 2.5, 1.5, 0.5},
Values: []float64{5, 4, 3, 2, 1, 0},
Timestamps: timestampsExpected,
}
resultExpected := []netstorage.Result{r}
@ -6714,7 +6714,7 @@ func TestExecSuccess(t *testing.T) {
q := `rate((2000-time())[100s:100s])`
r := netstorage.Result{
MetricName: metricNameExpected,
Values: []float64{0, 0, 6.5, 4.5, 2.5, 0.5},
Values: []float64{0, 0, 6, 4, 2, 0},
Timestamps: timestampsExpected,
}
resultExpected := []netstorage.Result{r}
@ -6725,7 +6725,7 @@ func TestExecSuccess(t *testing.T) {
q := `rate((2000-time())[100s:100s] offset 100s)`
r := netstorage.Result{
MetricName: metricNameExpected,
Values: []float64{0, 0, 3.5, 5.5, 3.5, 1.5},
Values: []float64{0, 0, 7, 5, 3, 1},
Timestamps: timestampsExpected,
}
resultExpected := []netstorage.Result{r}
@ -6736,7 +6736,7 @@ func TestExecSuccess(t *testing.T) {
q := `rate((2000-time())[100s:100s] offset 100s)[:] offset 100s`
r := netstorage.Result{
MetricName: metricNameExpected,
Values: []float64{0, 0, 0, 3.5, 5.5, 3.5},
Values: []float64{0, 0, 0, 7, 5, 3},
Timestamps: timestampsExpected,
}
resultExpected := []netstorage.Result{r}

View file

@ -145,11 +145,11 @@ var rollupAggrFuncs = map[string]rollupFunc{
"zscore_over_time": rollupZScoreOverTime,
}
// VictoriaMetrics can increase lookbehind window in square brackets for these functions
// if the given window doesn't contain enough samples for calculations.
// VictoriaMetrics can extends lookbehind window for these functions
// in order to make sure it contains enough points for returning non-empty results.
//
// This is needed in order to return the expected non-empty graphs when zooming in the graph in Grafana,
// which is built with `func_name(metric[$__interval])` query.
// This is needed for returning the expected non-empty graphs when zooming in the graph in Grafana,
// which is built with `func_name(metric)` query.
var rollupFuncsCanAdjustWindow = map[string]bool{
"default_rollup": true,
"deriv": true,
@ -584,15 +584,23 @@ func (rc *rollupConfig) doInternal(dstValues []float64, tsm *timeseriesMap, valu
window := rc.Window
if window <= 0 {
window = rc.Step
if rc.MayAdjustWindow && window < maxPrevInterval {
// Adjust lookbehind window only if it isn't set explicilty, e.g. rate(foo).
// In the case of missing lookbehind window it should be adjusted in order to return non-empty graph
// when the window doesn't cover at least two raw samples (this is what most users expect).
//
// If the user explicitly sets the lookbehind window to some fixed value, e.g. rate(foo[1s]),
// then it is expected he knows what he is doing. Do not adjust the lookbehind window then.
//
// See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3483
window = maxPrevInterval
}
if rc.isDefaultRollup && rc.LookbackDelta > 0 && window > rc.LookbackDelta {
// Implicit window exceeds -search.maxStalenessInterval, so limit it to -search.maxStalenessInterval
// according to https://github.com/VictoriaMetrics/VictoriaMetrics/issues/784
window = rc.LookbackDelta
}
}
if rc.MayAdjustWindow && window < maxPrevInterval {
window = maxPrevInterval
}
rfa := getRollupFuncArg()
rfa.idx = 0
rfa.window = window

View file

@ -15,6 +15,10 @@ The following tip changes can be tested by building VictoriaMetrics components f
## tip
**Update note 1:** This and newer releases of VictoriaMetrics may return gaps for `rate(m[d])` queries on short time ranges if `[d]` lookbehind window is set expliticly. For example, `rate(http_requests_total[$__interval])`. This reduces confusion level when the user expects the needed results from the query with explicitly set lookbehind window. See [this issue](https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3483). The previous gap filling behaviour can be restored by removing explicit lookbehind window `[d]` from the query, e.g. by substituting the `rate(m[d])` with `rate(m)`. See [these docs](https://docs.victoriametrics.com/MetricsQL.html#implicit-query-conversions) for details.
* BUGFIX: [MetricsQL](https://docs.victoriametrics.com/MetricsQL.html): never extend explicitly set lookbehind window for [rate()](https://docs.victoriametrics.com/MetricsQL.html#rate) function. This reduces the level of confusion when the user expects the needed results after explicitly seting the lookbehind window `[d]` in the query `rate(m[d])`. Previously VictoriaMetrics could silently extend the lookbehind window, so it covers at least two raw samples. Now this behavior works only if the lookbehind window in square brackets isn't set explicitly, e.g. in the case of `rate(m)`. See [this issue](https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3483) for details.
## [v1.85.2](https://github.com/VictoriaMetrics/VictoriaMetrics/releases/tag/v1.85.2)