Propagate ComputeStatus to deployment workflow (#11273)

## What changed?
Added a missing `d.syncSummary()` call in
`syncVersionDataToComputeStatus`, so it now notifies the parent
Deployment workflow after pulling a compute status from WCI.

## Why?
Without this, the pull only updates the Version workflow's own state.
The Deployment workflow (which
`ListWorkerDeployments`/`DescribeWorkerDeployment` actually read from)
is not updated, so `computeStatus` can stay permanently missing from the
API even when the data is available.

## How did you test it?
- [x] built
- [x] run locally and tested manually
- [ ] covered by existing tests
- [ ] added new unit test(s)
- [ ] added new functional test(s)

config/dynamicconfig/development-sql.yaml

@@ -40,6 +40,20 @@
 #matching.wv.VersionDrainageStatusRefreshInterval:
 # - value: 5s
 #
+### Compute config (Worker Controller Instance) coverage for the replay tester.
+### Required for the compute-config section of replaytester/worker/worker.go; without
+### `workercontroller.enabled` the CreateWorkerDeploymentVersion call with a compute
+### config is rejected and the version workflow never gets a non-nil ComputeConfig.
+#
+#workercontroller.enabled:
+# - value: true
+#workercontroller.compute_providers.enabled:
+# - value: ["test-invoke", "subprocess"] # leave unset to allow every registered provider
+#
+### Leave workercontroller.periodic_validation_interval_s at its 6h default. The driver populates
+### ComputeStatus by asking the WCI to re-validate its stored spec on demand, so it does not need
+### the periodic timer.
+#
 ### END of Worker Versioning Replay Test configs
 
 limit.maxIDLength:

go.mod

@@ -67,7 +67,7 @@ require (
 	go.opentelemetry.io/otel/sdk/metric v1.43.0
 	go.opentelemetry.io/otel/trace v1.44.0
 	go.temporal.io/api v1.63.5-0.20260803183639-0e1e8c485f37
-	go.temporal.io/auto-scaled-workers v0.0.0-20260706201056-4320b34799ee
+	go.temporal.io/auto-scaled-workers v0.0.0-20260728174133-979694be4176
 	go.temporal.io/sdk v1.44.0
 	go.uber.org/fx v1.24.0
 	go.uber.org/goleak v1.3.0

go.sum

@@ -481,8 +481,8 @@ go.opentelemetry.io/proto/slim/otlp/profiles/v1development v0.3.0 h1:CQvJSldHRUN
 go.opentelemetry.io/proto/slim/otlp/profiles/v1development v0.3.0/go.mod h1:xSQ+mEfJe/GjK1LXEyVOoSI1N9JV9ZI923X5kup43W4=
 go.temporal.io/api v1.63.5-0.20260803183639-0e1e8c485f37 h1:ZDICI5Hxc97YsjpE6/WREWkUqQ7qq+ntTH+IbOuTxNw=
 go.temporal.io/api v1.63.5-0.20260803183639-0e1e8c485f37/go.mod h1:SrlW2JMwVlDP4nRWSNznUFqnSHd+YeMDS1BkYo63HCQ=
-go.temporal.io/auto-scaled-workers v0.0.0-20260706201056-4320b34799ee h1:y6A65Iml06cR3CpxW2Zn8FQLjniPDTnN4jtr69UZxXI=
-go.temporal.io/auto-scaled-workers v0.0.0-20260706201056-4320b34799ee/go.mod h1:hhHijO9XRPIkAflLJJHix61M9FzbRPqk8fSydkcLkqw=
+go.temporal.io/auto-scaled-workers v0.0.0-20260728174133-979694be4176 h1:6AUglT8D3HsbOmV2zwwPpQOmxFXZN4NrMe6Z69n3fSc=
+go.temporal.io/auto-scaled-workers v0.0.0-20260728174133-979694be4176/go.mod h1:hhHijO9XRPIkAflLJJHix61M9FzbRPqk8fSydkcLkqw=
 go.temporal.io/sdk v1.44.0 h1:suitPDukX74rW3/N1FqvEbZTZVJJsxMKhv0KMa/j7pU=
 go.temporal.io/sdk v1.44.0/go.mod h1:vkApR12F9/Y8OR+hkxe7WyXQFuCX6clhzqnAk6rzDAM=
 go.uber.org/atomic v1.5.0/go.mod h1:sABNBOSYdrvTF6hTgEIbc7YasKWGhgEQZyfxyTvoXHQ=

service/worker/workerdeployment/replaytester/generate_history.sh

@@ -9,6 +9,10 @@
 # matching.PollerHistoryTTL=1s
 # matching.wv.VersionDrainageStatusVisibilityGracePeriod=5s
 # matching.wv.VersionDrainageStatusRefreshInterval=5s
+# workercontroller.enabled=true
+#
+# workercontroller.enabled is required by the compute-config section of worker/worker.go
+# (exerciseComputeConfig)
 #
 # Make sure you set deploymentWorkflowVersion correctly. It should be an integer instead of {WORKFLOW_VERSION} above.
 # If deploymentWorkflowVersion is set to i, run this script with v{i} as the argument. The argument instructs the script
@@ -25,6 +29,8 @@
 
 deploymentName="foo"
 version="1.0"
+# The compute-config-only version (created by exerciseComputeConfig in worker/worker.go)
+computeConfigVersion="2.0"
 
 # Expected workflow counts - users can override these if their changes are expected to generate more workflows which will be true when a breaking change to 
 # these workflows is introduced.
@@ -66,7 +72,7 @@ download_workflow_chain() {
     # Use the working query method with TemporalNamespaceDivision
     echo "   Getting the chain of CAN runs for this workflow using the TemporalNamespaceDivision query..."
     run_ids=$(temporal workflow list \
-        --query "TemporalNamespaceDivision = \"TemporalWorkerDeployment\" AND WorkflowType = \"$workflow_type\"" \
+        --query "TemporalNamespaceDivision = \"TemporalWorkerDeployment\" AND WorkflowType = \"$workflow_type\" AND WorkflowId = \"$workflow_id\"" \
         --output json | \
         jq -r '.[] | .execution.runId')
 
@@ -89,11 +95,19 @@ download_workflow_chain() {
         if [ -n "$run_id" ]; then
             echo "   Downloading run $((run_index + 1))/$run_count: $run_id"
             
-            temporal workflow show \
-                -w "$workflow_id" \
-                -r "$run_id" \
-                --output json | \
-                gzip -9c > "$run_dir/replay_${workflow_name}_run_${run_id}.json.gz"
+            # Write to a plain file first so the exit status is `workflow show`'s rather than
+            # gzip's, then check it is non-empty.
+            json_file="$run_dir/replay_${workflow_name}_run_${run_id}.json"
+            if ! temporal workflow show -w "$workflow_id" -r "$run_id" --output json > "$json_file"; then
+                echo "   Failed to download history for $workflow_id run $run_id" >&2
+                exit 1
+            fi
+            if [ ! -s "$json_file" ]; then
+                echo "   Empty history downloaded for $workflow_id run $run_id" >&2
+                exit 1
+            fi
+            gzip -9c < "$json_file" > "$json_file.gz"
+            rm -f "$json_file"
 
             ((run_index++))
         fi
@@ -115,6 +129,7 @@ echo "📁 Creating run directory: $run_dir"
 # Download all executions for both workflow types
 download_workflow_chain "temporal-sys-worker-deployment:$deploymentName" "worker_deployment_wf" "temporal-sys-worker-deployment-workflow" "$run_dir"
 download_workflow_chain "temporal-sys-worker-deployment-version:$deploymentName:$version" "worker_deployment_version_wf" "temporal-sys-worker-deployment-version-workflow" "$run_dir"
+download_workflow_chain "temporal-sys-worker-deployment-version:$deploymentName:$computeConfigVersion" "worker_deployment_version_wf" "temporal-sys-worker-deployment-version-workflow" "$run_dir"
 
 echo ""
 echo "🎉 Complete! All workflow execution histories downloaded to $run_dir"
@@ -140,3 +155,17 @@ ACTUAL_VERSION_WORKFLOWS=$version_files
 EOF
 
 echo "   📝 Expected counts saved to: $run_dir/expected_counts.txt"
+
+# A run whose actual counts don't match the expected ones would be checked in with an
+# expected_counts.txt that fails TestReplays immediately, so fail here instead.
+if [ "$deployment_files" -ne "$EXPECTED_DEPLOYMENT_WORKFLOWS" ] || [ "$version_files" -ne "$EXPECTED_VERSION_WORKFLOWS" ]; then
+  echo "" >&2
+  echo "❌ Workflow counts do not match the expected values." >&2
+  echo "   Expected: Deployment=$EXPECTED_DEPLOYMENT_WORKFLOWS, Version=$EXPECTED_VERSION_WORKFLOWS" >&2
+  echo "   Actual:   Deployment=$deployment_files, Version=$version_files" >&2
+  echo "" >&2
+  echo "   If the new counts are correct for your change, re-run with:" >&2
+  echo "     EXPECTED_DEPLOYMENT_WORKFLOWS=$deployment_files EXPECTED_VERSION_WORKFLOWS=$version_files $0 $1" >&2
+  echo "   Otherwise your change created extra workflow executions; investigate before checking in." >&2
+  exit 1
+fi

service/worker/workerdeployment/replaytester/testdata/v2/run_1785380013/expected_counts.txt

@@ -0,0 +1,6 @@
+# Expected workflow counts for replay testing
+# Generated by generate_history.sh on Wed Jul 29 19:53:35 PDT 2026
+EXPECTED_DEPLOYMENT_WORKFLOWS=23
+EXPECTED_VERSION_WORKFLOWS=14
+ACTUAL_DEPLOYMENT_WORKFLOWS=23
+ACTUAL_VERSION_WORKFLOWS=14

service/worker/workerdeployment/replaytester/testdata/v2/run_1785380306/expected_counts.txt

@@ -0,0 +1,6 @@
+# Expected workflow counts for replay testing
+# Generated by generate_history.sh on Wed Jul 29 19:58:28 PDT 2026
+EXPECTED_DEPLOYMENT_WORKFLOWS=24
+EXPECTED_VERSION_WORKFLOWS=14
+ACTUAL_DEPLOYMENT_WORKFLOWS=24
+ACTUAL_VERSION_WORKFLOWS=14

service/worker/workerdeployment/replaytester/worker/worker.go

@@ -5,13 +5,39 @@ import (
 	"log"
 	"time"
 
+	commonpb "go.temporal.io/api/common/v1"
+	computepb "go.temporal.io/api/compute/v1"
+	deploymentpb "go.temporal.io/api/deployment/v1"
+	enumspb "go.temporal.io/api/enums/v1"
 	"go.temporal.io/api/workflowservice/v1"
 	"go.temporal.io/sdk/client"
+	"go.temporal.io/sdk/converter"
 	"go.temporal.io/sdk/worker"
 	"go.temporal.io/sdk/workflow"
 	"go.temporal.io/server/service/worker/workerdeployment/replaytester"
 )
 
+// computeProviderType is the WCI compute provider used for the compute-config part of this
+// driver.
+const computeProviderType = "test-invoke"
+
+// computeScalerType specifies the scaler type for the compute configuration
+const computeScalerType = "no-sync"
+
+// newScalingGroup builds a scaling group backed by the local test provider.
+func newScalingGroup(note string, taskTypes ...enumspb.TaskQueueType) *computepb.ComputeConfigScalingGroup {
+	return &computepb.ComputeConfigScalingGroup{
+		TaskQueueTypes: taskTypes,
+		Provider: &computepb.ComputeProvider{
+			Type:    computeProviderType,
+			Details: encodeProviderDetails(map[string]any{"note": note}),
+		},
+		Scaler: &computepb.ComputeScaler{
+			Type: computeScalerType,
+		},
+	}
+}
+
 func main() {
 	// The client and worker are heavyweight objects that should be created once per process.
 	c, err := client.Dial(client.Options{})
@@ -23,6 +49,8 @@ func main() {
 	identity := "test-identity"
 	deploymentName := "foo"
 	build1 := "1.0"
+	// build2 is never polled by a worker; it exists only to contain a compute config.
+	build2 := "2.0"
 	v1 := worker.WorkerDeploymentVersion{
 		DeploymentName: deploymentName,
 		BuildID:        build1,
@@ -183,6 +211,10 @@ func main() {
 	// Make sure 1.0 is drained.
 	verifyDeployment(dHandle, "", "", 0, client.WorkerDeploymentVersionDrainageStatusDrained)
 
+	// Exercise the compute config (Worker Controller Instance) paths. Requires
+	// `workercontroller.enabled: true` in config/dynamicconfig/development-sql.yaml.
+	exerciseComputeConfig(c, deploymentName, build2, identity)
+
 	// Stopping both workers
 	w1.Stop()
 	w2.Stop()
@@ -200,6 +232,17 @@ func main() {
 		log.Fatalf("Unable to delete version: %v", err)
 	}
 
+	// Delete the compute-config version. This also tears down its Worker Controller
+	// Instance from inside the version workflow.
+	_, err = dHandle.DeleteVersion(context.Background(), client.WorkerDeploymentDeleteVersionOptions{
+		BuildID:      build2,
+		SkipDrainage: true,
+		Identity:     identity,
+	})
+	if err != nil {
+		log.Fatalf("Unable to delete version %s: %v", build2, err)
+	}
+
 	// Delete the deployment
 	_, err = deploymentClient.Delete(context.Background(), client.WorkerDeploymentDeleteOptions{
 		Name:     deploymentName,
@@ -211,6 +254,151 @@ func main() {
 
 }
 
+// exerciseComputeConfig drives the compute config paths of the version and deployment
+// workflows so the captured histories cover them.
+//
+// In the version workflow this covers:
+//   - starting with a non-nil VersionLocalState.ComputeConfig, which satisfies the
+//     `ComputeConfig != nil` guard.
+//   - the SignalSyncValidationStatus handler, which sets ComputeStatus and then signals the
+//     deployment workflow via syncSummary.
+//   - handleUpdateVersionComputeConfig, which also calls syncSummary.
+//   - ValidateComputeConfig, both as a dry-run of proposed changes and as a re-validation of
+//     the stored spec.
+//   - the DescribeWorkerControllerInstanceStatus pull inside syncVersionDataToComputeStatus
+//     returning a non-nil ProviderValidation
+//
+// In the deployment workflow this covers handleCreateWorkerDeploymentVersion's WCI creation
+// and the SyncVersionSummary signals arriving from the version workflow.
+//
+// After regenerating, confirm the new version-workflow history still contains a
+// "sync-compute-status-to-deployment" Version marker followed by a sync-version-summary signal;
+// if it does not, that fixture does not cover the pull path.
+//
+//nolint:revive
+func exerciseComputeConfig(c client.Client, deploymentName, buildID, identity string) {
+	version := &deploymentpb.WorkerDeploymentVersion{
+		DeploymentName: deploymentName,
+		BuildId:        buildID,
+	}
+
+	// Create the version with a compute config. This creates the WCI first, then starts the
+	// version workflow with the resulting ComputeConfigSummary in its initial state.
+	_, err := c.WorkflowService().CreateWorkerDeploymentVersion(context.Background(), &workflowservice.CreateWorkerDeploymentVersionRequest{
+		Namespace:         client.DefaultNamespace,
+		DeploymentVersion: version,
+		ComputeConfig: &computepb.ComputeConfig{
+			ScalingGroups: map[string]*computepb.ComputeConfigScalingGroup{
+				"group1": newScalingGroup("replaytester-group1", enumspb.TASK_QUEUE_TYPE_WORKFLOW),
+			},
+		},
+		Identity:  identity,
+		RequestId: "replaytester-create-version-" + buildID,
+	})
+	if err != nil {
+		log.Fatalf("Unable to create version %s with a compute config (is workercontroller.enabled set?): %v", buildID, err)
+	}
+
+	// Dry-run validation of a proposed scaling group: exercises ValidateComputeConfig without
+	// mutating the stored spec. Does not set ValidationStatus.
+	_, err = c.WorkflowService().ValidateWorkerDeploymentVersionComputeConfig(context.Background(), &workflowservice.ValidateWorkerDeploymentVersionComputeConfigRequest{
+		Namespace:         client.DefaultNamespace,
+		DeploymentVersion: version,
+		ComputeConfigScalingGroups: map[string]*computepb.ComputeConfigScalingGroupUpdate{
+			"dry-run-group": {
+				ScalingGroup: newScalingGroup("replaytester-dry-run", enumspb.TASK_QUEUE_TYPE_ACTIVITY),
+			},
+		},
+		Identity: identity,
+	})
+	if err != nil {
+		log.Fatalf("Unable to dry-run validate compute config for version %s: %v", buildID, err)
+	}
+
+	// Re-validate the stored spec by sending no changes. This is the path that sets
+	// ValidationStatus on the WCI and signals the version workflow, which sets ComputeStatus and
+	// calls syncSummary.
+	_, err = c.WorkflowService().ValidateWorkerDeploymentVersionComputeConfig(context.Background(), &workflowservice.ValidateWorkerDeploymentVersionComputeConfigRequest{
+		Namespace:         client.DefaultNamespace,
+		DeploymentVersion: version,
+		Identity:          identity,
+	})
+	if err != nil {
+		log.Fatalf("Unable to re-validate the stored compute config for version %s "+
+			"(does the pinned temporal-auto-scaled-workers include PR #99?): %v", buildID, err)
+	}
+
+	// Wait for the validation signal to land, so the captured history contains the ComputeStatus
+	// update and the syncSummary it triggers.
+	waitForComputeStatus(c, version)
+
+	// Add a second scaling group, exercising handleUpdateVersionComputeConfig.
+	_, err = c.WorkflowService().UpdateWorkerDeploymentVersionComputeConfig(context.Background(), &workflowservice.UpdateWorkerDeploymentVersionComputeConfigRequest{
+		Namespace:         client.DefaultNamespace,
+		DeploymentVersion: version,
+		ComputeConfigScalingGroups: map[string]*computepb.ComputeConfigScalingGroupUpdate{
+			"group2": {
+				ScalingGroup: newScalingGroup("replaytester-group2", enumspb.TASK_QUEUE_TYPE_ACTIVITY),
+			},
+		},
+		Identity:  identity,
+		RequestId: "replaytester-update-compute-config-" + buildID,
+	})
+	if err != nil {
+		log.Fatalf("Unable to update compute config for version %s: %v", buildID, err)
+	}
+
+	// Remove the group we just added, exercising the removal branch of the same handler.
+	_, err = c.WorkflowService().UpdateWorkerDeploymentVersionComputeConfig(context.Background(), &workflowservice.UpdateWorkerDeploymentVersionComputeConfigRequest{
+		Namespace:                        client.DefaultNamespace,
+		DeploymentVersion:                version,
+		RemoveComputeConfigScalingGroups: []string{"group2"},
+		Identity:                         identity,
+		RequestId:                        "replaytester-remove-compute-config-" + buildID,
+	})
+	if err != nil {
+		log.Fatalf("Unable to remove compute config scaling group for version %s: %v", buildID, err)
+	}
+}
+
+// waitForComputeStatus polls DescribeWorkerDeployment until the version's ComputeStatus shows
+// up in the deployment workflow's version summary.
+//
+//nolint:revive
+func waitForComputeStatus(c client.Client, version *deploymentpb.WorkerDeploymentVersion) {
+	log.Printf("Waiting for compute status to propagate to the deployment workflow for version %s...", version.GetBuildId())
+	for range 40 {
+		resp, err := c.WorkflowService().DescribeWorkerDeployment(context.Background(), &workflowservice.DescribeWorkerDeploymentRequest{
+			Namespace:      client.DefaultNamespace,
+			DeploymentName: version.GetDeploymentName(),
+		})
+		if err != nil {
+			log.Fatalf("Unable to describe deployment %s: %v", version.GetDeploymentName(), err)
+		}
+		for _, summary := range resp.GetWorkerDeploymentInfo().GetVersionSummaries() {
+			if summary.GetDeploymentVersion().GetBuildId() != version.GetBuildId() {
+				continue
+			}
+			if summary.GetComputeStatus().GetProviderValidation() != nil {
+				return
+			}
+		}
+		time.Sleep(500 * time.Millisecond)
+	}
+	log.Fatalf("Timed out waiting for compute status to propagate to the deployment workflow for version %s", version.GetBuildId())
+}
+
+// encodeProviderDetails encodes provider-specific config the way the server decodes it
+//
+//nolint:revive
+func encodeProviderDetails(details map[string]any) *commonpb.Payload {
+	p, err := converter.GetDefaultDataConverter().ToPayload(details)
+	if err != nil {
+		log.Fatalf("Unable to encode compute provider details: %v", err)
+	}
+	return p
+}
+
 //nolint:revive
 func verifyDeployment(dHandle client.WorkerDeploymentHandle,
 	expectedCurrentVersionBuildId string,

service/worker/workerdeployment/version_workflow.go

@@ -1268,6 +1268,9 @@ func (d *VersionWorkflowRunner) syncVersionDataToComputeStatus(ctx workflow.Cont
 				logger.Error("failed to sync compute status", "error", err)
 			} else if result.ProviderValidation != nil {
 				state.ComputeStatus = &result
+				if workflow.GetVersion(ctx, "sync-compute-status-to-deployment", workflow.DefaultVersion, 0) >= 0 {
+					d.syncSummary(ctx) // propagate updated ComputeStatus to deployment workflow
+				}
 			}
 		})
 	}

tests/worker_deployment_version_test.go

@@ -1189,6 +1189,12 @@ func (s *DeploymentVersionSuite) TestUpdateVersionMetadata() {
 	s.Equal(metadataIdentity, resp.GetWorkerDeploymentVersionInfo().GetLastModifierIdentity())
 }
 
+// testInvokeScaler returns the scaler that scaling groups backed by the test-invoke compute
+// provider must declare.
+func testInvokeScaler() *computepb.ComputeScaler {
+	return &computepb.ComputeScaler{Type: "no-sync"}
+}
+
 func (s *DeploymentVersionSuite) createDeploymentAndVersion(
 	env *testcore.TestEnv,
 	tv *testvars.TestVars,
@@ -1228,7 +1234,7 @@ func (s *DeploymentVersionSuite) TestUpdateComputeConfig_Success() {
 
 	s.createDeploymentAndVersion(env, env.Tv(), createIdentity, &computepb.ComputeConfig{
 		ScalingGroups: map[string]*computepb.ComputeConfigScalingGroup{
-			"sg1": {Provider: validProvider},
+			"sg1": {Provider: validProvider, Scaler: testInvokeScaler()},
 		},
 	})
 
@@ -1242,6 +1248,7 @@ func (s *DeploymentVersionSuite) TestUpdateComputeConfig_Success() {
 				ScalingGroup: &computepb.ComputeConfigScalingGroup{
 					TaskQueueTypes: []enumspb.TaskQueueType{enumspb.TASK_QUEUE_TYPE_ACTIVITY},
 					Provider:       validProvider,
+					Scaler:         testInvokeScaler(),
 				},
 			},
 		},
@@ -1259,10 +1266,11 @@ func (s *DeploymentVersionSuite) TestUpdateComputeConfig_Success() {
 		a.Equal(updateIdentity, info.GetLastModifierIdentity())
 		a.True(proto.Equal(&computepb.ComputeConfig{
 			ScalingGroups: map[string]*computepb.ComputeConfigScalingGroup{
-				"sg1": {Provider: validProvider},
+				"sg1": {Provider: validProvider, Scaler: testInvokeScaler()},
 				"sg2": {
 					TaskQueueTypes: []enumspb.TaskQueueType{enumspb.TASK_QUEUE_TYPE_ACTIVITY},
 					Provider:       validProvider,
+					Scaler:         testInvokeScaler(),
 				},
 			},
 		}, info.GetComputeConfig()))
@@ -1335,6 +1343,7 @@ func (s *DeploymentVersionSuite) TestUpdateComputeConfig_UpdateExistingGroup() {
 			"sg1": {
 				TaskQueueTypes: []enumspb.TaskQueueType{enumspb.TASK_QUEUE_TYPE_WORKFLOW},
 				Provider:       validProvider,
+				Scaler:         testInvokeScaler(),
 			},
 		},
 	})
@@ -1366,6 +1375,7 @@ func (s *DeploymentVersionSuite) TestUpdateComputeConfig_UpdateExistingGroup() {
 				"sg1": {
 					TaskQueueTypes: []enumspb.TaskQueueType{enumspb.TASK_QUEUE_TYPE_ACTIVITY},
 					Provider:       validProvider,
+					Scaler:         testInvokeScaler(),
 				},
 			},
 		}, descResp.GetWorkerDeploymentVersionInfo().GetComputeConfig()))
@@ -1378,10 +1388,11 @@ func (s *DeploymentVersionSuite) TestUpdateComputeConfig_RemoveScalingGroup() {
 
 	s.createDeploymentAndVersion(env, env.Tv(), env.Tv().Any().String(), &computepb.ComputeConfig{
 		ScalingGroups: map[string]*computepb.ComputeConfigScalingGroup{
-			"sg1": {Provider: validProvider},
+			"sg1": {Provider: validProvider, Scaler: testInvokeScaler()},
 			"sg2": {
 				TaskQueueTypes: []enumspb.TaskQueueType{enumspb.TASK_QUEUE_TYPE_ACTIVITY},
 				Provider:       validProvider,
+				Scaler:         testInvokeScaler(),
 			},
 		},
 	})
@@ -1406,6 +1417,7 @@ func (s *DeploymentVersionSuite) TestUpdateComputeConfig_RemoveScalingGroup() {
 				"sg2": {
 					TaskQueueTypes: []enumspb.TaskQueueType{enumspb.TASK_QUEUE_TYPE_ACTIVITY},
 					Provider:       validProvider,
+					Scaler:         testInvokeScaler(),
 				},
 			},
 		}, descResp.GetWorkerDeploymentVersionInfo().GetComputeConfig()))
@@ -1445,7 +1457,7 @@ func (s *DeploymentVersionSuite) TestUpdateComputeConfig_InvalidProvider() {
 	env := s.newTestEnv()
 	s.createDeploymentAndVersion(env, env.Tv(), env.Tv().Any().String(), &computepb.ComputeConfig{
 		ScalingGroups: map[string]*computepb.ComputeConfigScalingGroup{
-			"sg1": {Provider: computeprovider.TestInvokeComputeProviderValidComputeProvider()},
+			"sg1": {Provider: computeprovider.TestInvokeComputeProviderValidComputeProvider(), Scaler: testInvokeScaler()},
 		},
 	})
 
@@ -1529,6 +1541,7 @@ func (s *DeploymentVersionSuite) TestValidateComputeConfig_Valid() {
 			"sg1": {
 				ScalingGroup: &computepb.ComputeConfigScalingGroup{
 					Provider: computeprovider.TestInvokeComputeProviderValidComputeProvider(),
+					Scaler:   testInvokeScaler(),
 				},
 			},
 		},
@@ -1583,6 +1596,7 @@ func (s *DeploymentVersionSuite) TestValidateComputeConfig_VersionNotFound() {
 			"sg1": {
 				ScalingGroup: &computepb.ComputeConfigScalingGroup{
 					Provider: computeprovider.TestInvokeComputeProviderValidComputeProvider(),
+					Scaler:   testInvokeScaler(),
 				},
 			},
 		},
@@ -3093,6 +3107,7 @@ func (s *DeploymentVersionSuite) TestCreateWorkerDeploymentVersion_Success() {
 		ScalingGroups: map[string]*computepb.ComputeConfigScalingGroup{
 			"sg1": {
 				Provider: computeprovider.TestInvokeComputeProviderValidComputeProvider(),
+				Scaler:   testInvokeScaler(),
 			},
 		},
 	}
@@ -3413,6 +3428,7 @@ func (s *DeploymentVersionSuite) TestCreateWorkerDeploymentVersion_MultipleVersi
 		ScalingGroups: map[string]*computepb.ComputeConfigScalingGroup{
 			"sg1": {
 				Provider: computeprovider.TestInvokeComputeProviderValidComputeProvider(),
+				Scaler:   testInvokeScaler(),
 			},
 		},
 	}
@@ -3420,6 +3436,7 @@ func (s *DeploymentVersionSuite) TestCreateWorkerDeploymentVersion_MultipleVersi
 		ScalingGroups: map[string]*computepb.ComputeConfigScalingGroup{
 			"sg2": {
 				Provider: computeprovider.TestInvokeComputeProviderValidComputeProvider(),
+				Scaler:   testInvokeScaler(),
 			},
 		},
 	}