QueryService¶

API Version: e6data.io/v1alpha1 Kind: QueryService Short Names: qs, cluster

1. Purpose¶

QueryService deploys and manages the query execution components of an e6data analytics cluster. This includes:

Planner: Query parsing, optimization, and execution planning
Queue: Query queuing, scheduling, and cluster coordination
Executor: Distributed query execution workers
Transpiler (optional): SQL dialect translation sidecar
Gateway (optional): External API gateway for programmatic access

Note: Traffic routing is now handled by TrafficInfra which deploys Envoy + xDS for dynamic load balancing. HAProxy has been removed from QueryService.

Create a QueryService after NamespaceConfig and MetadataServices are running. Each QueryService represents a logical analytics cluster that users connect to for running queries.

Note: Infrastructure settings (cloud, tolerations, node selectors, image pull secrets) are now managed by NamespaceConfig. QueryService inherits these settings automatically.

2. High-level Behavior¶

When you create a QueryService CR, the operator:

Inherits infrastructure settings from NamespaceConfig in the same namespace
Discovers MetadataServices in the same namespace for metadata endpoints
Deploys all components with blue-green deployment strategy
Creates Services for internal and external access
Configures autoscaling if enabled (executor scaling)
Integrates with Pool for burst capacity (if poolRef specified)
Manages auto-suspension/resume if configured

Prerequisites¶

NamespaceConfig must exist in the same namespace (provides cloud, storage backend, scheduling config)
MetadataServices must exist in the same namespace (provides storage/schema endpoints)

Child Resources Created¶

Resource Type	Name Pattern	Purpose
Deployment	`{name}-planner-{blue\\|green}`	Query planner pods
Deployment	`{name}-queue-{blue\\|green}`	Queue/coordinator pods
Deployment	`{name}-executor-{blue\\|green}`	Query executor pods
Deployment	`{name}-executor-pool-{blue\\|green}`	Pool executor pods (if poolRef)
Deployment	`{name}-gateway`	Gateway pods (if enabled)
ConfigMap	`{name}-{component}-config-{blue\\|green}`	Component configurations
Service	`{name}-planner-{blue\\|green}`	Headless planner services (for xDS discovery)
Service	`{name}-queue-{blue\\|green}`	Queue services
Service	`{name}-gateway`	Gateway service (if enabled)

Note: Planner services are now headless (ClusterIP: None) with a GRPC port (9001) for xDS endpoint discovery by TrafficInfra.

External Dependencies¶

MetadataServices: Must exist in same namespace (provides storage/schema endpoints)
TrafficInfra (required for external access): Provides Envoy + xDS for traffic routing
Pool (optional): For burst capacity beyond regular nodes
Object Storage: Access to data lake for query execution

3. Spec Reference¶

3.1 Top-level Fields¶

Field	Type	Required	Default	Description
`alias`	string	Yes	-	Cluster alias (human-readable identifier)
`workspace`	string	Yes	-	Workspace name (must match MetadataServices)
`clusterUUID`	string	No	CR name	Unique cluster identifier
`baseDomain`	string	No	-	Base domain for endpoints
`planner`	PlannerSpec	Yes	-	Planner configuration
`queue`	QueueSpec	Yes	-	Queue configuration
`executor`	ExecutorSpec	Yes	-	Executor configuration
`transpiler`	TranspilerSpec	No	Disabled	Transpiler sidecar
`gateway`	GatewaySpec	No	Disabled	Gateway deployment
`autoSuspension`	AutoSuspensionSpec	No	Disabled	Auto-suspend configuration
`autoResume`	AutoResumeSpec	No	Disabled	Auto-resume configuration
`queryHistory`	QueryHistorySpec	No	Disabled	Query event logging
`podAnnotations`	map[string]string	No	`{}`	Annotations for all pods

Inherited from NamespaceConfig¶

The following fields are inherited from NamespaceConfig and no longer specified in QueryService:

Field	Description
`cloud`	Cloud provider (AWS, GCP, AZURE)
`storageBackend`	Object storage path (s3a://, gs://, abfs://)
`s3Endpoint`	Custom S3 endpoint for S3-compatible storage
`imageRepository`	Container registry path
`imagePullSecrets`	Secrets for private registries
`tolerations`	Pod tolerations for scheduling
`nodeSelector`	Node labels for pod placement
`affinity`	Advanced scheduling rules
`karpenterNodePool`	Karpenter NodePool name
`serviceAccount`	ServiceAccount for pods (via `serviceAccounts.data`)
`defaultPoolRef`	Default pool for burst capacity

3.2 Planner (PlannerSpec)¶

Field	Type	Required	Default	Description
`replicas`	int32	No	`1`	Number of planner pods
`image`	ImageSpec	Yes	-	Planner image configuration
`resources`	ResourceSpec	Yes	-	CPU/Memory limits
`ports.web`	int32	No	`10001`	Web/query port
`ports.grpc`	int32	No	`9001`	GRPC port (for Envoy xDS routing)
`ports.metrics`	int32	No	`8081`	Metrics port
`queryTimeoutSeconds`	int64	No	`900`	Query timeout (15 min default)
`jwksEnabled`	bool	No	`false`	Enable JWKS authentication
`configVariables`	map[string]string	No	`{}`	config.properties entries (see ConfigVariables Reference)
`environmentVariables`	map[string]string	No	`{}`	Container env vars (see EnvironmentVariables Reference)
`extraConfig`	map[string]string	No	`{}`	Additional config properties

3.3 Queue (QueueSpec)¶

Field	Type	Required	Default	Description
`image`	ImageSpec	Yes	-	Queue image configuration
`resources`	ResourceSpec	Yes	-	CPU/Memory limits
`ports.cache`	int32	No	`9007`	Cache port
`ports.clusterManager`	int32	No	`9004`	Cluster manager port
`ports.thrift`	int32	No	`9003`	Thrift RPC port
`ports.web`	int32	No	`10003`	HTTP API port
`ports.metrics`	int32	No	`8081`	Metrics port
`configVariables`	map[string]string	No	`{}`	config.properties entries (see ConfigVariables Reference)
`environmentVariables`	map[string]string	No	`{}`	Container env vars (see EnvironmentVariables Reference)

3.4 Executor (ExecutorSpec)¶

Field	Type	Required	Default	Description
`replicas`	int32	No	`1`	Number of executor pods
`image`	ExecutorImageSpec	Yes	-	Executor image (tag required)
`resources`	ResourceSpec	Yes	-	CPU/Memory limits
`ports.thrift`	int32	No	`9002`	Thrift port
`ports.web`	int32	No	`10002`	Health check port
`ports.metrics`	int32	No	`8081`	Metrics port
`health`	HealthConfig	No	See defaults	Health probe config
`configVariables`	map[string]string	No	`{}`	config.properties entries (see ConfigVariables Reference)
`environmentVariables`	map[string]string	No	`{}`	Container env vars (see EnvironmentVariables Reference)
`affinity`	Affinity	No	-	Pod affinity rules
`autoscaling`	AutoscalingSpec	No	Disabled	Executor autoscaling
`poolRef`	PoolReference	No	-	Pool for burst capacity

3.5 Transpiler (TranspilerSpec)¶

Field	Type	Required	Default	Description
`enabled`	bool	No	`false`	Enable transpiler sidecar
`image`	ImageSpec	No	-	Transpiler image
`resources`	ResourceSpec	No	-	CPU/Memory limits
`ports.http`	int32	No	`8100`	HTTP port
`environmentVariables`	map[string]string	No	`{}`	Container env vars

Note: Transpiler runs as a sidecar in the planner pod, not as a separate deployment.

3.6 Gateway (GatewaySpec)¶

Field	Type	Required	Default	Description
`enabled`	bool	No	`false`	Enable gateway deployment
`replicas`	int32	No	`1`	Number of gateway pods
`image`	ImageSpec	No	-	Gateway image
`resources`	ResourceSpec	No	-	CPU/Memory limits
`ports.http`	int32	No	`8083`	HTTP port
`ports.metrics`	int32	No	`8081`	Metrics port
`configVariables`	map[string]string	No	`{}`	config.properties entries (see ConfigVariables Reference)
`environmentVariables`	map[string]string	No	`{}`	Container env vars (see EnvironmentVariables Reference)
`service.type`	string	No	`NodePort`	Service type
`service.port`	int32	No	`8083`	Service port

Note: Gateway is a standalone deployment (no blue-green). ENGINE_HOST/ENGINE_PORT are auto-configured.

3.7 AutoSuspension¶

Field	Type	Required	Default	Description
`enabled`	bool	No	`false`	Enable auto-suspension
`maxIdleDurationMinutes`	int32	Conditional	-	Idle timeout before suspend

3.8 AutoResume¶

Field	Type	Required	Default	Description
`enabled`	bool	No	`false`	Enable auto-resume

When enabled, the cluster automatically resumes on incoming requests via the operator API.

3.9 QueryHistory¶

Field	Type	Required	Default	Description
`enabled`	bool	No	`true`	Enable query history collection
`image`	QueryHistoryImageSpec	No	`timberio/vector:0.41.0-alpine`	Vector sidecar image
`s3Prefix`	string	No	`query-history`	S3 prefix for logs
`s3Partition`	string	No	`hour`	Partition by `hour` or `day`
`httpPort`	int32	No	`2020`	Vector HTTP source port
`timeoutSeconds`	int32	No	`5`	Publish timeout
`messageQueueCapacity`	int32	No	`10000`	Internal buffer size
`numPublishingThreads`	int32	No	`10`	Concurrent publishers
`maxPayloadBatchSize`	int32	No	`100`	Events per batch
`batchTimeoutSecs`	int32	No	`30`	S3 batch timeout
`batchMaxBytes`	int64	No	`10485760`	S3 batch size limit
`compression`	string	No	`gzip`	Output compression
`resources`	ResourceSpec	No	-	Vector sidecar resources
`greptimeRef`	QueryHistoryGreptimeRef	No	-	GreptimeDB for real-time queries

3.10 Autoscaling¶

Field	Type	Required	Default	Description
`enabled`	bool	No	`false`	Enable autoscaling
`minExecutors`	int32	No	Current replicas	Minimum executors
`maxExecutors`	int32	Conditional	-	Maximum executors
`clusterManagementBaseURL`	string	No	Auto-generated	Operator API URL
`windowBased.enabled`	bool	No	`false`	Enable window-based scaling
`windowBased.slidingWindowDuration`	int32	Conditional	-	Window duration (seconds)

3.11 PoolReference¶

Field	Type	Required	Default	Description
`name`	string	Yes	-	Pool name
`namespace`	string	No	Same namespace	Pool namespace

3.12 ConfigVariables Reference¶

ConfigVariables are written to a config.properties file mounted in the container. These are application-level settings that configure the e6data services.

Auto-Populated Variables (Do Not Override)¶

The operator automatically sets these values. Specifying them in configVariables will be ignored:

Variable	Auto-Value	Description
`CLOUD`	Auto-detected	Cloud provider (AWS/GCP/AZURE)
`ALIAS`	`spec.alias`	Cluster alias
`WORKSPACE`	`spec.workspace`	Workspace name
`CLUSTER_UUID`	`spec.clusterUUID`	Cluster identifier

Common Planner ConfigVariables¶

Variable	Type	Default	Description
`ENABLE_COST_BASED_OPTIMIZATION`	bool	`true`	Enable cost-based query optimization
`MAX_SPLIT_SIZE_BYTES`	int	`134217728`	Maximum split size (128MB)
`ENABLE_DYNAMIC_FILTERING`	bool	`true`	Enable runtime filtering
`ENABLE_QUERY_CACHING`	bool	`false`	Cache query results
`MAX_CACHED_QUERY_AGE_SECONDS`	int	`3600`	Query cache TTL

Common Queue ConfigVariables¶

Variable	Type	Default	Description
`MAX_CONCURRENT_QUERIES`	int	`100`	Max concurrent queries
`QUERY_QUEUE_SIZE`	int	`1000`	Query queue buffer size
`ENABLE_QUERY_PRIORITIES`	bool	`true`	Enable query priority levels
`DEFAULT_QUERY_PRIORITY`	int	`5`	Default priority (1-10)

Common Executor ConfigVariables¶

Variable	Type	Default	Description
`SPILL_ENABLED`	bool	`true`	Enable spilling to disk
`SPILL_PATH`	string	`/nvme/spill`	Spill directory path
`MAX_MEMORY_PER_QUERY_BYTES`	int	Auto	Memory limit per query
`ENABLE_BLOOM_FILTERS`	bool	`true`	Use bloom filters

Common Gateway ConfigVariables¶

Variable	Type	Default	Description
`ENGINE_HOST`	string	Auto	Envoy proxy host (auto-configured)
`ENGINE_PORT`	int	Auto	Envoy proxy port (auto-configured)
`API_VERSION`	string	`v1`	API version prefix

3.13 EnvironmentVariables Reference¶

EnvironmentVariables are set as container environment variables. These control runtime behavior.

Auto-Populated Variables (Do Not Override)¶

The operator automatically sets these values from pod metadata:

Variable	Auto-Value	Description
`IS_KUBE`	`"true"`	Indicates Kubernetes environment
`POD_NAME`	Pod metadata	Current pod name
`POD_IP`	Pod status	Current pod IP
`NAMESPACE`	Pod metadata	Current namespace
`JAVA_TOOL_OPTIONS`	Auto-calculated	JVM options (80% of memory)

Common EnvironmentVariables¶

Variable	Type	Default	Description
`E6_LOGGING_LEVEL`	string	`E6_INFO`	Log level: `E6_DEBUG`, `E6_INFO`, `E6_WARN`, `E6_ERROR`
`LOG_FORMAT`	string	`json`	Log format: `json`, `text`
`TZ`	string	`UTC`	Timezone
`ENABLE_JMX`	bool	`true`	Enable JMX metrics
`JMX_PORT`	int	`9010`	JMX port

Example: Custom Configuration¶

planner:
  configVariables:
    ENABLE_COST_BASED_OPTIMIZATION: "true"
    MAX_SPLIT_SIZE_BYTES: "268435456"  # 256MB
  environmentVariables:
    E6_LOGGING_LEVEL: "E6_DEBUG"
    TZ: "America/New_York"

queue:
  configVariables:
    MAX_CONCURRENT_QUERIES: "50"
    ENABLE_QUERY_PRIORITIES: "true"
  environmentVariables:
    E6_LOGGING_LEVEL: "E6_INFO"

executor:
  configVariables:
    SPILL_ENABLED: "true"
    ENABLE_BLOOM_FILTERS: "true"
  environmentVariables:
    E6_LOGGING_LEVEL: "E6_INFO"

4. Example Manifests¶

4.1 Minimal Example¶

apiVersion: e6data.io/v1alpha1
kind: QueryService
metadata:
  name: analytics-cluster
  namespace: workspace-analytics-prod
spec:
  alias: analytics
  workspace: analytics-prod

  planner:
    image:
      tag: "3.0.217"
    resources:
      memory: "4Gi"
      cpu: "2"

  queue:
    image:
      tag: "3.0.217"
    resources:
      memory: "4Gi"
      cpu: "2"

  executor:
    replicas: 2
    image:
      tag: "3.0.217"
    resources:
      memory: "30Gi"
      cpu: "15"

Note: For external access, create a TrafficInfra CR that references the planner services.

4.2 Production Example (Full Configuration)¶

apiVersion: e6data.io/v1alpha1
kind: QueryService
metadata:
  name: analytics-cluster
  namespace: workspace-analytics-prod
  labels:
    e6data.io/workspace: analytics-prod
    e6data.io/environment: production
    e6data.io/pool: burst-pool  # For pool selector matching
spec:
  alias: analytics
  workspace: analytics-prod
  clusterUUID: analytics-cluster-prod-001
  baseDomain: e6data.example.com

  # Planner - Query Optimization
  planner:
    replicas: 2
    image:
      tag: "3.0.217"
    resources:
      memory: "8Gi"
      cpu: "4"
    queryTimeoutSeconds: 1800  # 30 minutes
    jwksEnabled: true
    configVariables:
      ENABLE_COST_BASED_OPTIMIZATION: "true"

  # Queue - Query Scheduling
  queue:
    replicas: 2
    image:
      tag: "3.0.217"
    resources:
      memory: "8Gi"
      cpu: "4"
    configVariables:
      MAX_CONCURRENT_QUERIES: "50"

  # Executor - Query Workers
  executor:
    replicas: 4
    image:
      tag: "3.0.217"
    resources:
      memory: "60Gi"
      cpu: "30"
    health:
      initialDelaySeconds: 30
      periodSeconds: 10
      failureThreshold: 30

    # Autoscaling configuration
    autoscaling:
      enabled: true
      minExecutors: 2
      maxExecutors: 20
      windowBased:
        enabled: true
        slidingWindowDuration: 300  # 5 minutes

    # Pool for burst capacity
    poolRef:
      name: burst-pool
      namespace: e6-pools

  # Transpiler - SQL Dialect Translation (optional)
  transpiler:
    enabled: true
    image:
      tag: "1.0.0"
    resources:
      memory: "1Gi"
      cpu: "500m"

  # Gateway - External API (optional)
  gateway:
    enabled: true
    replicas: 2
    image:
      tag: "1.0.0"
    resources:
      memory: "1Gi"
      cpu: "500m"
    service:
      type: LoadBalancer
      port: 8083

  # Auto-suspension for cost savings
  autoSuspension:
    enabled: true
    maxIdleDurationMinutes: 30

  autoResume:
    enabled: true

  # Query history logging
  queryHistory:
    enabled: true
    s3Prefix: "query-history/analytics-cluster"
    s3Partition: hour
    compression: gzip
    greptimeRef:
      name: greptime-cluster
      database: public
      table: query_history

  # Scheduling
  tolerations:
    - key: "e6data-workspace-name"
      operator: "Equal"
      value: "analytics-prod"
      effect: "NoSchedule"

  karpenterNodePool: query-services

  podAnnotations:
    prometheus.io/scrape: "true"
    prometheus.io/port: "8081"

4.3 Development/Testing Example¶

apiVersion: e6data.io/v1alpha1
kind: QueryService
metadata:
  name: dev-cluster
  namespace: workspace-dev
spec:
  alias: dev
  workspace: dev

  planner:
    replicas: 1
    image:
      tag: "3.0.217"
    resources:
      memory: "2Gi"
      cpu: "1"
    queryTimeoutSeconds: 300  # 5 minutes for dev

  queue:
    replicas: 1
    image:
      tag: "3.0.217"
    resources:
      memory: "2Gi"
      cpu: "1"

  executor:
    replicas: 1
    image:
      tag: "3.0.217"
    resources:
      memory: "8Gi"
      cpu: "4"

  # Auto-suspend after 10 minutes idle
  autoSuspension:
    enabled: true
    maxIdleDurationMinutes: 10

  autoResume:
    enabled: true

5. Status & Lifecycle¶

5.1 Status Fields¶

Field	Type	Description
`phase`	string	Current lifecycle phase
`message`	string	Human-readable status message
`ready`	bool	`true` when all components ready
`activeStrategy`	string	Current active deployment (`blue` or `green`)
`activeReleaseVersion`	string	Currently running version
`pendingStrategy`	string	Deployment being prepared
`deploymentPhase`	string	Blue-green phase: `Stable`, `Deploying`, `Switching`, `Draining`, `Cleanup`
`plannerDeployment`	DeploymentStatus	Planner status
`queueDeployment`	DeploymentStatus	Queue status
`executorDeployment`	DeploymentStatus	Executor status
`poolExecutorDeployment`	DeploymentStatus	Pool executor status
`transpilerDeployment`	DeploymentStatus	Transpiler status
`gatewayDeployment`	DeploymentStatus	Gateway status
`plannerService`	string	Planner service endpoint
`queueService`	string	Queue service endpoint
`gatewayService`	string	Gateway service endpoint
`poolName`	string	Attached pool name
`poolNamespace`	string	Attached pool namespace
`regularExecutorReplicas`	int32	Executors on regular nodes
`poolExecutorReplicas`	int32	Executors on pool nodes
`scalingHistory`	[]ScalingEvent	Last 20 scaling operations
`suspensionHistory`	[]SuspensionEvent	Last 20 suspend/resume events
`releaseHistory`	[]QueryReleaseRecord	Last 10 releases

5.2 Phase Values¶

Phase	Description
`Waiting`	Waiting for dependencies (MetadataServices)
`Pending`	Resource created, waiting to reconcile
`Creating`	Initial deployment in progress
`Deploying`	Blue-green deployment in progress
`Ready`	All components healthy
`Running`	Alias for Ready
`Updating`	Configuration update in progress
`Failed`	Deployment failed
`Degraded`	Partially healthy

5.3 Scaling Events¶

The operator tracks scaling operations:

scalingHistory:
  - timestamp: "2024-01-15T10:30:00Z"
    component: executor
    oldReplicas: 4
    newReplicas: 8
    trigger: autoscaling-api
    strategy: blue

5.4 Suspension Events¶

suspensionHistory:
  - timestamp: "2024-01-15T22:00:00Z"
    action: suspend
    trigger: auto-suspension-api
    strategy: blue
    componentsSuspended: [executor, planner, queue]
    preSuspensionReplicas:
      executorReplicas: 4
      plannerReplicas: 2
      queueReplicas: 1

Dependencies¶

CRD	Relationship
NamespaceConfig	Required - provides cloud, storage, scheduling configuration
MetadataServices	Required - must exist in same namespace for metadata endpoints
TrafficInfra	Required for external access - provides Envoy + xDS traffic routing
Pool	Optional - for burst capacity via `poolRef`
GreptimeDBCluster	Optional - for real-time query history

Labels Applied¶

app.kubernetes.io/name: {planner|queue|executor}
app.kubernetes.io/instance: {cr-name}
app.kubernetes.io/component: {component}
app.kubernetes.io/managed-by: e6-operator
e6data.io/workspace: {workspace}
e6data.io/cluster: {alias}
e6data.io/strategy: {blue|green}

7. Troubleshooting¶

7.1 Common Issues¶

Cluster Stuck in "Waiting" Phase¶

Symptoms:

$ kubectl get qs
NAME                PHASE     STRATEGY
analytics-cluster   Waiting

Cause: MetadataServices not ready in same namespace.

Fix:

# Check MetadataServices status
kubectl get mds -n workspace-analytics-prod

# Verify MetadataServices is Ready
kubectl get mds analytics-prod -o jsonpath='{.status.phase}'

Executors Not Scaling¶

Symptoms: Executor count doesn't change despite autoscaling enabled.

Checks:

# Verify autoscaling config
kubectl get qs analytics-cluster -o jsonpath='{.spec.executor.autoscaling}'

# Check operator logs for scaling decisions
kubectl logs -n e6-operator-system -l app=e6-operator | grep -i scaling

# Check scaling history
kubectl get qs analytics-cluster -o jsonpath='{.status.scalingHistory}'

Pool Executors Not Starting¶

Symptoms: poolExecutorReplicas: 0 despite pool reference.

Checks:

# Verify pool exists and is active
kubectl get pool burst-pool -n e6-pools

# Check pool allocations
kubectl get pool burst-pool -o jsonpath='{.status.allocations}'

# Verify QueryService is in pool's allowed list
kubectl get pool burst-pool -o jsonpath='{.spec.allowedQueryServices}'

Auto-Resume Not Working¶

Symptoms: Cluster doesn't wake up on incoming queries.

Checks:

# Verify auto-resume is enabled
kubectl get qs analytics-cluster -o jsonpath='{.spec.autoResume.enabled}'

# Check Envoy proxy is running (handles traffic routing)
kubectl get pods -l e6data.io/component=envoy

# Check xDS control plane logs
kubectl logs -l app=xds-control-plane

7.2 Useful Commands¶

# Get QueryService status
kubectl get qs analytics-cluster -o yaml

# Watch all components
kubectl get pods -l app.kubernetes.io/instance=analytics-cluster -w

# Check component replicas
kubectl get deploy -l e6data.io/cluster=analytics

# View scaling history
kubectl get qs analytics-cluster -o jsonpath='{.status.scalingHistory[*]}' | jq

# Manual scale (for testing)
kubectl patch qs analytics-cluster --type=merge -p '{"spec":{"executor":{"replicas":8}}}'

# Trigger manual suspend (via annotation)
kubectl annotate qs analytics-cluster e6data.io/suspend=true

# Trigger manual resume
kubectl annotate qs analytics-cluster e6data.io/suspend-

# Check blue-green deployment status
kubectl get deploy -l e6data.io/strategy=blue
kubectl get deploy -l e6data.io/strategy=green

8. Autoscaling API¶

The operator exposes an HTTP API for autoscaling at:

http://e6-operator.{namespace}.svc.cluster.local:8082/endpoints/v1/cluster/{namespace}/{name}/autoscale

Request:

POST /endpoints/v1/cluster/workspace-analytics-prod/analytics-cluster/autoscale
{
  "desired_executors": 10
}

Response:

{
  "status": "scaling",
  "current_executors": 4,
  "target_executors": 10,
  "operation_id": "scale-xyz123"
}

The API automatically splits scaling between regular and pool executors based on minExecutors configuration.

QueryService¶

1. Purpose¶

2. High-level Behavior¶

Prerequisites¶

Child Resources Created¶

External Dependencies¶

3. Spec Reference¶

3.1 Top-level Fields¶

Inherited from NamespaceConfig¶

3.2 Planner (PlannerSpec)¶

3.3 Queue (QueueSpec)¶

3.4 Executor (ExecutorSpec)¶

3.5 Transpiler (TranspilerSpec)¶

3.6 Gateway (GatewaySpec)¶

3.7 AutoSuspension¶

3.8 AutoResume¶

3.9 QueryHistory¶

3.10 Autoscaling¶

3.11 PoolReference¶

3.12 ConfigVariables Reference¶

Auto-Populated Variables (Do Not Override)¶

Common Planner ConfigVariables¶

Common Queue ConfigVariables¶

Common Executor ConfigVariables¶

Common Gateway ConfigVariables¶

3.13 EnvironmentVariables Reference¶

Auto-Populated Variables (Do Not Override)¶

Common EnvironmentVariables¶

Example: Custom Configuration¶

4. Example Manifests¶

4.1 Minimal Example¶

4.2 Production Example (Full Configuration)¶

4.3 Development/Testing Example¶

5. Status & Lifecycle¶

5.1 Status Fields¶

5.2 Phase Values¶

5.3 Scaling Events¶

5.4 Suspension Events¶

6. Related Resources¶

Dependencies¶

Labels Applied¶

7. Troubleshooting¶

7.1 Common Issues¶

Cluster Stuck in "Waiting" Phase¶

Executors Not Scaling¶

Pool Executors Not Starting¶

Auto-Resume Not Working¶

7.2 Useful Commands¶

8. Autoscaling API¶