vmsingle and vmagent: keep_input for streaming aggregation

VictoriaMetrics · Aug 16, 2024 · 5a82e0e · 5a82e0e
1 parent 64879fb
commit 5a82e0e
Show file tree

Hide file tree

Showing 5 changed files with 23 additions and 0 deletions.
diff --git a/api/operator/v1beta1/vmextra_types.go b/api/operator/v1beta1/vmextra_types.go
@@ -563,6 +563,9 @@ type StreamAggrRule struct {
 	// +optional
 	IgnoreOldSamples *bool `json:"ignore_old_samples,omitempty" yaml:"ignore_old_samples,omitempty"`
 
+	// KeepInput instructs whether to keep all the matched input samples after the aggregation
+        KeepInput *bool `json:"keep_input,omitempty" yaml:"keep_input,omitempty"`
+
 	// By is an optional list of labels for grouping input series.
 	//
 	// See also Without.

diff --git a/api/operator/v1beta1/zz_generated.deepcopy.go b/api/operator/v1beta1/zz_generated.deepcopy.go
diff --git a/config/crd/overlay/crd.yaml b/config/crd/overlay/crd.yaml
@@ -3291,6 +3291,10 @@ spec:
                               interval:
                                 description: Interval is the interval between aggregations.
                                 type: string
+                              keep_input:
+                                description: KeepInput instructs whether to keep all
+                                  the matched input samples after the aggregation
+                                type: boolean
                               keep_metric_names:
                                 description: KeepMetricNames instructs to leave metric
                                   names as is for the output time series without adding
@@ -5090,6 +5094,10 @@ spec:
                         interval:
                           description: Interval is the interval between aggregations.
                           type: string
+                        keep_input:
+                          description: KeepInput instructs whether to keep all the
+                            matched input samples after the aggregation
+                          type: boolean
                         keep_metric_names:
                           description: KeepMetricNames instructs to leave metric names
                             as is for the output time series without adding any suffix.
@@ -28534,6 +28542,10 @@ spec:
                         interval:
                           description: Interval is the interval between aggregations.
                           type: string
+                        keep_input:
+                          description: KeepInput instructs whether to keep all the
+                            matched input samples after the aggregation
+                          type: boolean
                         keep_metric_names:
                           description: KeepMetricNames instructs to leave metric names
                             as is for the output time series without adding any suffix.

diff --git a/docs/CHANGELOG.md b/docs/CHANGELOG.md
@@ -75,6 +75,8 @@ aliases:
 
 - [vmscrapeconfig](https://docs.victoriametrics.com/operator/api#vmscrapeconfig) - added `max_scrape_size` parameter for scrape protocols configuration
 
+- [vmagent](./api.md#vmagent) and [vmsingle](./api.md#vmsingle): adds new fields into `streamAggrConfig`: `keep_input`.
+
 <a name="v0.45.0"></a>
 
 ## [v0.45.0](https://github.com/VictoriaMetrics/operator/releases/tag/v0.45.0) - 10 Jun 2024

diff --git a/docs/api.md b/docs/api.md
@@ -1857,6 +1857,7 @@ _Appears in:_
 | `ignore_old_samples` | IgnoreOldSamples instructs to ignore samples with old timestamps outside the current aggregation interval. | _boolean_ | false |
 | `input_relabel_configs` | InputRelabelConfigs is an optional relabeling rules, which are applied on the input<br />before aggregation. | _[RelabelConfig](#relabelconfig) array_ | false |
 | `interval` | Interval is the interval between aggregations. | _string_ | true |
+| `keep_input` | KeepInput instructs whether to keep all the matched input samples after the aggregation | _boolean_ | true |
 | `keep_metric_names` | KeepMetricNames instructs to leave metric names as is for the output time series without adding any suffix. | _boolean_ | false |
 | `match` | Match is a label selector (or list of label selectors) for filtering time series for the given selector.<br /><br />If the match isn't set, then all the input time series are processed. | _[StringOrArray](#stringorarray)_ | false |
 | `no_align_flush_to_interval` | NoAlignFlushToInterval disables aligning of flushes to multiples of Interval.<br />By default flushes are aligned to Interval. | _boolean_ | false |