first claude draft

proto/metrics.proto

@@ -0,0 +1,159 @@
+syntax = "proto3";
+
+package monitoring;
+
+option go_package = "github.com/your-org/sysmonstm/proto";
+
+// MetricsService handles streaming metrics from collectors to aggregator
+service MetricsService {
+  // Client-side streaming: collector streams metrics to aggregator
+  rpc StreamMetrics(stream Metric) returns (StreamAck) {}
+
+  // Get current state of a machine
+  rpc GetCurrentState(StateRequest) returns (MachineState) {}
+
+  // Get current state of all machines
+  rpc GetAllStates(Empty) returns (AllMachinesState) {}
+}
+
+// ControlService handles bidirectional control commands
+service ControlService {
+  // Bidirectional streaming for commands and responses
+  rpc Control(stream ControlCommand) returns (stream ControlResponse) {}
+}
+
+// ConfigService handles dynamic configuration
+service ConfigService {
+  // Get current configuration for a collector
+  rpc GetConfig(ConfigRequest) returns (CollectorConfig) {}
+
+  // Stream configuration updates
+  rpc WatchConfig(ConfigRequest) returns (stream CollectorConfig) {}
+}
+
+// Empty message for requests with no parameters
+message Empty {}
+
+// Basic metric message
+message Metric {
+  string machine_id = 1;
+  string hostname = 2;
+  int64 timestamp_ms = 3;
+  MetricType type = 4;
+  double value = 5;
+  map<string, string> labels = 6;
+}
+
+// Batch of metrics for efficient transmission
+message MetricBatch {
+  string machine_id = 1;
+  string hostname = 2;
+  int64 timestamp_ms = 3;
+  repeated MetricPoint metrics = 4;
+}
+
+message MetricPoint {
+  MetricType type = 1;
+  double value = 2;
+  map<string, string> labels = 3;
+}
+
+enum MetricType {
+  METRIC_TYPE_UNSPECIFIED = 0;
+  CPU_PERCENT = 1;
+  CPU_PERCENT_PER_CORE = 2;
+  MEMORY_PERCENT = 3;
+  MEMORY_USED_BYTES = 4;
+  MEMORY_AVAILABLE_BYTES = 5;
+  DISK_PERCENT = 6;
+  DISK_USED_BYTES = 7;
+  DISK_READ_BYTES_SEC = 8;
+  DISK_WRITE_BYTES_SEC = 9;
+  NETWORK_SENT_BYTES_SEC = 10;
+  NETWORK_RECV_BYTES_SEC = 11;
+  NETWORK_CONNECTIONS = 12;
+  PROCESS_COUNT = 13;
+  LOAD_AVG_1M = 14;
+  LOAD_AVG_5M = 15;
+  LOAD_AVG_15M = 16;
+}
+
+// Acknowledgment for streamed metrics
+message StreamAck {
+  bool success = 1;
+  int64 metrics_received = 2;
+  string message = 3;
+}
+
+// Request for machine state
+message StateRequest {
+  string machine_id = 1;
+}
+
+// Current state of a single machine
+message MachineState {
+  string machine_id = 1;
+  string hostname = 2;
+  int64 last_seen_ms = 3;
+  repeated Metric current_metrics = 4;
+  HealthStatus health = 5;
+  map<string, string> metadata = 6;
+}
+
+// State of all machines
+message AllMachinesState {
+  repeated MachineState machines = 1;
+}
+
+enum HealthStatus {
+  HEALTH_STATUS_UNSPECIFIED = 0;
+  HEALTHY = 1;
+  WARNING = 2;
+  CRITICAL = 3;
+  UNKNOWN = 4;
+  OFFLINE = 5;
+}
+
+// Control commands for collectors
+message ControlCommand {
+  string command_id = 1;
+  oneof command {
+    UpdateIntervalCommand update_interval = 2;
+    RestartCollectionCommand restart = 3;
+    ShutdownCommand shutdown = 4;
+  }
+}
+
+message UpdateIntervalCommand {
+  int32 interval_seconds = 1;
+}
+
+message RestartCollectionCommand {}
+
+message ShutdownCommand {
+  bool graceful = 1;
+}
+
+message ControlResponse {
+  string command_id = 1;
+  bool success = 2;
+  string message = 3;
+}
+
+// Configuration messages
+message ConfigRequest {
+  string machine_id = 1;
+}
+
+message CollectorConfig {
+  int32 collection_interval_seconds = 1;
+  repeated MetricType enabled_metrics = 2;
+  map<string, string> labels = 3;
+  repeated ThresholdConfig thresholds = 4;
+}
+
+message ThresholdConfig {
+  MetricType metric_type = 1;
+  double warning_threshold = 2;
+  double critical_threshold = 3;
+}