HPA stabilizes initial recommendation

HPA will treat initial size of autoscalee to avoid hastily overriding
recomendations made by HPA (if HPA set size and then was restarted) or by user
(initial size should be treated as human-generated recommendation).

pkg/controller/podautoscaler/horizontal.go

@@ -464,6 +464,12 @@ func (a *HorizontalController) computeStatusForExternalMetric(currentReplicas in
 	return 0, time.Time{}, "", fmt.Errorf(errMsg)
 }
 
+func (a *HorizontalController) recordInitialRecommendation(currentReplicas int32, key string) {
+	if a.recommendations[key] == nil {
+		a.recommendations[key] = []timestampedRecommendation{{currentReplicas, time.Now()}}
+	}
+}
+
 func (a *HorizontalController) reconcileAutoscaler(hpav1Shared *autoscalingv1.HorizontalPodAutoscaler, key string) error {
 	// make a copy so that we never mutate the shared informer cache (conversion can mutate the object)
 	hpav1 := hpav1Shared.DeepCopy()
@@ -508,6 +514,7 @@ func (a *HorizontalController) reconcileAutoscaler(hpav1Shared *autoscalingv1.Ho
 	}
 	setCondition(hpa, autoscalingv2.AbleToScale, v1.ConditionTrue, "SucceededGetScale", "the HPA controller was able to get the target's current scale")
 	currentReplicas := scale.Status.Replicas
+	a.recordInitialRecommendation(currentReplicas, key)
 
 	var metricStatuses []autoscalingv2.MetricStatus
 	metricDesiredReplicas := int32(0)

pkg/controller/podautoscaler/horizontal_test.go

@@ -1112,6 +1112,32 @@ func TestScaleDown(t *testing.T) {
 		reportedLevels:          []uint64{100, 300, 500, 250, 250},
 		reportedCPURequests:     []resource.Quantity{resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0")},
 		useMetricsAPI:           true,
+		recommendations:         []timestampedRecommendation{},
+	}
+	tc.runTest(t)
+}
+
+func TestScaleDownStabilizeInitialSize(t *testing.T) {
+	tc := testCase{
+		minReplicas:             2,
+		maxReplicas:             6,
+		initialReplicas:         5,
+		expectedDesiredReplicas: 5,
+		CPUTarget:               50,
+		verifyCPUCurrent:        true,
+		reportedLevels:          []uint64{100, 300, 500, 250, 250},
+		reportedCPURequests:     []resource.Quantity{resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0")},
+		useMetricsAPI:           true,
+		recommendations:         nil,
+		expectedConditions: statusOkWithOverrides(autoscalingv2.HorizontalPodAutoscalerCondition{
+			Type:   autoscalingv2.AbleToScale,
+			Status: v1.ConditionTrue,
+			Reason: "ReadyForNewScale",
+		}, autoscalingv2.HorizontalPodAutoscalerCondition{
+			Type:   autoscalingv2.AbleToScale,
+			Status: v1.ConditionTrue,
+			Reason: "ScaleDownStabilized",
+		}),
 	}
 	tc.runTest(t)
 }
@@ -1139,6 +1165,7 @@ func TestScaleDownCM(t *testing.T) {
 		},
 		reportedLevels:      []uint64{12000, 12000, 12000, 12000, 12000},
 		reportedCPURequests: []resource.Quantity{resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0")},
+		recommendations:     []timestampedRecommendation{},
 	}
 	tc.runTest(t)
 }
@@ -1171,6 +1198,7 @@ func TestScaleDownCMObject(t *testing.T) {
 		},
 		reportedLevels:      []uint64{12000},
 		reportedCPURequests: []resource.Quantity{resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0")},
+		recommendations:     []timestampedRecommendation{},
 	}
 	tc.runTest(t)
 }
@@ -1195,7 +1223,8 @@ func TestScaleDownCMExternal(t *testing.T) {
 				},
 			},
 		},
-		reportedLevels: []uint64{8600},
+		reportedLevels:  []uint64{8600},
+		recommendations: []timestampedRecommendation{},
 	}
 	tc.runTest(t)
 }
@@ -1220,7 +1249,8 @@ func TestScaleDownPerPodCMExternal(t *testing.T) {
 				},
 			},
 		},
-		reportedLevels: []uint64{8600},
+		reportedLevels:  []uint64{8600},
+		recommendations: []timestampedRecommendation{},
 	}
 	tc.runTest(t)
 }
@@ -1238,6 +1268,7 @@ func TestScaleDownIncludeUnreadyPods(t *testing.T) {
 		reportedCPURequests:     []resource.Quantity{resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0")},
 		useMetricsAPI:           true,
 		reportedPodReadiness:    []v1.ConditionStatus{v1.ConditionTrue, v1.ConditionTrue, v1.ConditionTrue, v1.ConditionFalse, v1.ConditionFalse},
+		recommendations:         []timestampedRecommendation{},
 	}
 	tc.runTest(t)
 }
@@ -1255,6 +1286,7 @@ func TestScaleDownIgnoreHotCpuPods(t *testing.T) {
 		reportedCPURequests:     []resource.Quantity{resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0")},
 		useMetricsAPI:           true,
 		reportedPodStartTime:    []metav1.Time{coolCpuCreationTime(), coolCpuCreationTime(), coolCpuCreationTime(), hotCpuCreationTime(), hotCpuCreationTime()},
+		recommendations:         []timestampedRecommendation{},
 	}
 	tc.runTest(t)
 }
@@ -1273,6 +1305,7 @@ func TestScaleDownIgnoresFailedPods(t *testing.T) {
 		useMetricsAPI:           true,
 		reportedPodReadiness:    []v1.ConditionStatus{v1.ConditionTrue, v1.ConditionTrue, v1.ConditionTrue, v1.ConditionTrue, v1.ConditionTrue, v1.ConditionFalse, v1.ConditionFalse},
 		reportedPodPhase:        []v1.PodPhase{v1.PodRunning, v1.PodRunning, v1.PodRunning, v1.PodRunning, v1.PodRunning, v1.PodFailed, v1.PodFailed},
+		recommendations:         []timestampedRecommendation{},
 	}
 	tc.runTest(t)
 }
@@ -1292,6 +1325,7 @@ func TestScaleDownIgnoresDeletionPods(t *testing.T) {
 		reportedPodReadiness:         []v1.ConditionStatus{v1.ConditionTrue, v1.ConditionTrue, v1.ConditionTrue, v1.ConditionTrue, v1.ConditionTrue, v1.ConditionFalse, v1.ConditionFalse},
 		reportedPodPhase:             []v1.PodPhase{v1.PodRunning, v1.PodRunning, v1.PodRunning, v1.PodRunning, v1.PodRunning, v1.PodRunning, v1.PodRunning},
 		reportedPodDeletionTimestamp: []bool{false, false, false, false, false, true, true},
+		recommendations:              []timestampedRecommendation{},
 	}
 	tc.runTest(t)
 }
@@ -1457,6 +1491,7 @@ func TestMinReplicas(t *testing.T) {
 			Status: v1.ConditionTrue,
 			Reason: "TooFewReplicas",
 		}),
+		recommendations: []timestampedRecommendation{},
 	}
 	tc.runTest(t)
 }
@@ -1476,6 +1511,7 @@ func TestMinReplicasDesiredZero(t *testing.T) {
 			Status: v1.ConditionTrue,
 			Reason: "TooFewReplicas",
 		}),
+		recommendations: []timestampedRecommendation{},
 	}
 	tc.runTest(t)
 }
@@ -1580,6 +1616,7 @@ func TestMissingMetrics(t *testing.T) {
 		reportedLevels:          []uint64{400, 95},
 		reportedCPURequests:     []resource.Quantity{resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0")},
 		useMetricsAPI:           true,
+		recommendations:         []timestampedRecommendation{},
 	}
 	tc.runTest(t)
 }
@@ -1665,6 +1702,7 @@ func TestMissingReports(t *testing.T) {
 		reportedLevels:          []uint64{200},
 		reportedCPURequests:     []resource.Quantity{resource.MustParse("0.2")},
 		useMetricsAPI:           true,
+		recommendations:         []timestampedRecommendation{},
 	}
 	tc.runTest(t)
 }
@@ -2168,7 +2206,8 @@ func TestComputedToleranceAlgImplementation(t *testing.T) {
 			resource.MustParse(fmt.Sprint(perPodRequested) + "m"),
 			resource.MustParse(fmt.Sprint(perPodRequested) + "m"),
 		},
-		useMetricsAPI: true,
+		useMetricsAPI:   true,
+		recommendations: []timestampedRecommendation{},
 	}
 
 	tc.runTest(t)
@@ -2241,6 +2280,7 @@ func TestAvoidUncessaryUpdates(t *testing.T) {
 		reportedPodStartTime:    []metav1.Time{coolCpuCreationTime(), hotCpuCreationTime(), hotCpuCreationTime()},
 		useMetricsAPI:           true,
 		lastScaleTime:           &now,
+		recommendations:         []timestampedRecommendation{},
 	}
 	testClient, _, _, _, _ := tc.prepareTestClient(t)
 	tc.testClient = testClient

pkg/controller/podautoscaler/legacy_horizontal_test.go

@@ -98,7 +98,8 @@ type legacyTestCase struct {
 	resource *fakeResource
 
 	// Last scale time
-	lastScaleTime *metav1.Time
+	lastScaleTime   *metav1.Time
+	recommendations []timestampedRecommendation
 }
 
 // Needs to be called under a lock.
@@ -504,6 +505,10 @@ func (tc *legacyTestCase) runTest(t *testing.T) {
 	)
 	hpaController.hpaListerSynced = alwaysReady
 
+	if tc.recommendations != nil {
+		hpaController.recommendations["test-namespace/test-hpa"] = tc.recommendations
+	}
+
 	stop := make(chan struct{})
 	defer close(stop)
 	informerFactory.Start(stop)
@@ -689,6 +694,7 @@ func TestLegacyScaleDown(t *testing.T) {
 		reportedLevels:      []uint64{100, 300, 500, 250, 250},
 		reportedCPURequests: []resource.Quantity{resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0")},
 		useMetricsAPI:       true,
+		recommendations:     []timestampedRecommendation{},
 	}
 	tc.runTest(t)
 }
@@ -711,6 +717,7 @@ func TestLegacyScaleDownCM(t *testing.T) {
 		},
 		reportedLevels:      []uint64{12, 12, 12, 12, 12},
 		reportedCPURequests: []resource.Quantity{resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0")},
+		recommendations:     []timestampedRecommendation{},
 	}
 	tc.runTest(t)
 }
@@ -728,6 +735,7 @@ func TestLegacyScaleDownIgnoresUnreadyPods(t *testing.T) {
 		reportedCPURequests:  []resource.Quantity{resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0"), resource.MustParse("1.0")},
 		useMetricsAPI:        true,
 		reportedPodReadiness: []v1.ConditionStatus{v1.ConditionTrue, v1.ConditionTrue, v1.ConditionTrue, v1.ConditionFalse, v1.ConditionFalse},
+		recommendations:      []timestampedRecommendation{},
 	}
 	tc.runTest(t)
 }