feat: add processingStrategy field in health events structure

data-models/protobufs/health_event.proto

@@ -29,6 +29,14 @@ message HealthEvents {
   repeated HealthEvent events = 2;
 }
 
+// ProcessingStrategy defines how downstream modules should handle the event.
+// EXECUTE_REMEDIATION: normal behavior; downstream modules may update cluster state.
+// STORE_ONLY: observability-only behavior; event should be persisted/exported but should not modify cluster resources.
+enum ProcessingStrategy {
+  EXECUTE_REMEDIATION = 0;
+  STORE_ONLY = 1;
+}
+
 enum RecommendedAction {
   NONE = 0;
   COMPONENT_RESET = 2;
@@ -66,6 +74,7 @@ message HealthEvent {
   string nodeName = 13;
   BehaviourOverrides quarantineOverrides = 14;
   BehaviourOverrides drainOverrides = 15;
+  ProcessingStrategy processingStrategy = 16;
 }
 
 message BehaviourOverrides {

event-exporter/pkg/transformer/cloudevents.go

@@ -63,6 +63,7 @@ func ToCloudEvent(event *pb.HealthEvent, metadata map[string]string) (*CloudEven
 		"entitiesImpacted":   entities,
 		"generatedTimestamp": timestamp,
 		"nodeName":           event.NodeName,
+		"processingStrategy": event.ProcessingStrategy.String(),
 	}
 
 	if len(event.Metadata) > 0 {

event-exporter/pkg/transformer/cloudevents_test.go

@@ -66,6 +66,7 @@ func TestToCloudEvent(t *testing.T) {
 					Force: false,
 					Skip:  true,
 				},
+				ProcessingStrategy: pb.ProcessingStrategy_STORE_ONLY,
 			},
 			metadata: map[string]string{
 				"cluster":     "prod-cluster-1",
@@ -102,6 +103,9 @@ func TestToCloudEvent(t *testing.T) {
 				if healthEvent["recommendedAction"] != "RESTART_VM" {
 					t.Errorf("recommendedAction = %v, want %v", healthEvent["recommendedAction"], "RESTART_VM")
 				}
+				if healthEvent["processingStrategy"] != "STORE_ONLY" {
+					t.Errorf("processingStrategy = %v, want STORE_ONLY", healthEvent["processingStrategy"])
+				}
 
 				entities := healthEvent["entitiesImpacted"].([]map[string]any)
 				if len(entities) != 2 {

fault-quarantine/pkg/evaluator/rule_evaluator_test.go

@@ -260,6 +260,7 @@ func TestRoundTrip(t *testing.T) {
 			"nanos":   float64(eventTime.GetNanos()),
 		},
 		"nodeName":            "test-node",
+		"processingStrategy":  float64(0),
 		"quarantineOverrides": nil,
 		"drainOverrides":      nil,
 	}

fault-quarantine/pkg/initializer/init.go

@@ -63,7 +63,7 @@ func InitializeAll(ctx context.Context, params InitializationParams) (*Component
 	}
 
 	builder := client.GetPipelineBuilder()
-	pipeline := builder.BuildAllHealthEventInsertsPipeline()
+	pipeline := builder.BuildProcessableHealthEventInsertsPipeline()
 
 	var tomlCfg config.TomlConfig
 	if err := configmanager.LoadTOMLConfig(params.TomlConfigPath, &tomlCfg); err != nil {

health-monitors/gpu-health-monitor/gpu_health_monitor/protos/health_event_pb2.pyi

@@ -11,6 +11,11 @@ from typing import ClassVar as _ClassVar, Optional as _Optional, Union as _Union
 
 DESCRIPTOR: _descriptor.FileDescriptor
 
+class ProcessingStrategy(int, metaclass=_enum_type_wrapper.EnumTypeWrapper):
+    __slots__ = ()
+    EXECUTE_REMEDIATION: _ClassVar[ProcessingStrategy]
+    STORE_ONLY: _ClassVar[ProcessingStrategy]
+
 class RecommendedAction(int, metaclass=_enum_type_wrapper.EnumTypeWrapper):
     __slots__ = ()
     NONE: _ClassVar[RecommendedAction]
@@ -23,6 +28,8 @@ class RecommendedAction(int, metaclass=_enum_type_wrapper.EnumTypeWrapper):
     RUN_DCGMEUD: _ClassVar[RecommendedAction]
     UNKNOWN: _ClassVar[RecommendedAction]
 
+EXECUTE_REMEDIATION: ProcessingStrategy
+STORE_ONLY: ProcessingStrategy
 NONE: RecommendedAction
 COMPONENT_RESET: RecommendedAction
 CONTACT_SUPPORT: RecommendedAction
@@ -68,6 +75,7 @@ class HealthEvent(_message.Message):
         "nodeName",
         "quarantineOverrides",
         "drainOverrides",
+        "processingStrategy",
     )
 
     class MetadataEntry(_message.Message):
@@ -93,6 +101,7 @@ class HealthEvent(_message.Message):
     NODENAME_FIELD_NUMBER: _ClassVar[int]
     QUARANTINEOVERRIDES_FIELD_NUMBER: _ClassVar[int]
     DRAINOVERRIDES_FIELD_NUMBER: _ClassVar[int]
+    PROCESSINGSTRATEGY_FIELD_NUMBER: _ClassVar[int]
     version: int
     agent: str
     componentClass: str
@@ -108,6 +117,7 @@ class HealthEvent(_message.Message):
     nodeName: str
     quarantineOverrides: BehaviourOverrides
     drainOverrides: BehaviourOverrides
+    processingStrategy: ProcessingStrategy
     def __init__(
         self,
         version: _Optional[int] = ...,
@@ -125,6 +135,7 @@ class HealthEvent(_message.Message):
         nodeName: _Optional[str] = ...,
         quarantineOverrides: _Optional[_Union[BehaviourOverrides, _Mapping]] = ...,
         drainOverrides: _Optional[_Union[BehaviourOverrides, _Mapping]] = ...,
+        processingStrategy: _Optional[_Union[ProcessingStrategy, str]] = ...,
     ) -> None: ...
 
 class BehaviourOverrides(_message.Message):