claude-mpm 4.1.4__py3-none-any.whl → 4.1.6__py3-none-any.whl

claude_mpm/services/infrastructure/monitoring/process.py

@@ -0,0 +1,342 @@
+"""Process health monitoring service.
+
+Monitors individual process health including CPU, memory, file descriptors, and threads.
+"""
+
+from typing import List
+
+from claude_mpm.core.constants import ResourceLimits, TimeoutConfig
+
+from .base import BaseMonitoringService, HealthMetric, HealthStatus
+
+try:
+    import psutil
+
+    PSUTIL_AVAILABLE = True
+except ImportError:
+    PSUTIL_AVAILABLE = False
+    psutil = None
+
+
+class ProcessHealthService(BaseMonitoringService):
+    """Service for monitoring individual process health.
+
+    Monitors:
+    - Process existence and status
+    - Process CPU usage
+    - Process memory usage (RSS, VMS)
+    - File descriptor count (Unix)
+    - Thread count
+    """
+
+    def __init__(
+        self,
+        pid: int,
+        cpu_threshold: float = 80.0,
+        memory_threshold_mb: int = 500,
+        fd_threshold: int = 1000,
+    ):
+        """Initialize process health service.
+
+        Args:
+            pid: Process ID to monitor
+            cpu_threshold: CPU usage threshold as percentage
+            memory_threshold_mb: Memory usage threshold in MB
+            fd_threshold: File descriptor count threshold
+        """
+        super().__init__(f"ProcessHealth_{pid}")
+        self.pid = pid
+        self.cpu_threshold = cpu_threshold
+        self.memory_threshold_mb = memory_threshold_mb
+        self.fd_threshold = fd_threshold
+        self.process = None
+
+        if PSUTIL_AVAILABLE:
+            try:
+                self.process = psutil.Process(pid)
+            except psutil.NoSuchProcess:
+                self.logger.warning(f"Process {pid} not found for monitoring")
+
+    async def check_health(self) -> List[HealthMetric]:
+        """Check process health metrics."""
+        metrics = []
+
+        if not PSUTIL_AVAILABLE:
+            metrics.append(
+                HealthMetric(
+                    name="psutil_availability",
+                    value=False,
+                    status=HealthStatus.WARNING,
+                    message="psutil not available for process monitoring",
+                )
+            )
+            return metrics
+
+        if not self.process:
+            metrics.append(
+                HealthMetric(
+                    name="process_exists",
+                    value=False,
+                    status=HealthStatus.CRITICAL,
+                    message=f"Process {self.pid} not found",
+                )
+            )
+            return metrics
+
+        try:
+            # Check if process still exists
+            if not self.process.is_running():
+                metrics.append(
+                    HealthMetric(
+                        name="process_exists",
+                        value=False,
+                        status=HealthStatus.CRITICAL,
+                        message=f"Process {self.pid} is no longer running",
+                    )
+                )
+                return metrics
+
+            # Process status
+            metrics.extend(self._check_process_status())
+
+            # CPU usage
+            metrics.extend(self._check_cpu_usage())
+
+            # Memory usage
+            metrics.extend(self._check_memory_usage())
+
+            # File descriptors
+            metrics.extend(self._check_file_descriptors())
+
+            # Thread count
+            metrics.extend(self._check_thread_count())
+
+            # Process metadata
+            metrics.extend(self._check_process_metadata())
+
+        except psutil.NoSuchProcess:
+            metrics.append(
+                HealthMetric(
+                    name="process_exists",
+                    value=False,
+                    status=HealthStatus.CRITICAL,
+                    message=f"Process {self.pid} no longer exists",
+                )
+            )
+        except Exception as e:
+            self.logger.error(f"Error checking process health: {e}")
+            metrics.append(
+                HealthMetric(
+                    name="process_check_error",
+                    value=str(e),
+                    status=HealthStatus.UNKNOWN,
+                    message=f"Unexpected error during process health check: {e}",
+                )
+            )
+
+        return metrics
+
+    def _check_process_status(self) -> List[HealthMetric]:
+        """Check process status."""
+        metrics = []
+        try:
+            status = self.process.status()
+            process_healthy = status not in [
+                psutil.STATUS_ZOMBIE,
+                psutil.STATUS_DEAD,
+                psutil.STATUS_STOPPED,
+            ]
+            metrics.append(
+                HealthMetric(
+                    name="process_status",
+                    value=status,
+                    status=(
+                        HealthStatus.HEALTHY
+                        if process_healthy
+                        else HealthStatus.CRITICAL
+                    ),
+                    message=f"Process status: {status}",
+                )
+            )
+        except Exception as e:
+            metrics.append(
+                HealthMetric(
+                    name="process_status",
+                    value="unknown",
+                    status=HealthStatus.UNKNOWN,
+                    message=f"Failed to get process status: {e}",
+                )
+            )
+        return metrics
+
+    def _check_cpu_usage(self) -> List[HealthMetric]:
+        """Check CPU usage."""
+        metrics = []
+        try:
+            cpu_percent = self.process.cpu_percent(
+                interval=TimeoutConfig.CPU_SAMPLE_INTERVAL
+            )
+            cpu_status = HealthStatus.HEALTHY
+            if cpu_percent > self.cpu_threshold:
+                cpu_status = (
+                    HealthStatus.WARNING
+                    if cpu_percent < self.cpu_threshold * 1.2
+                    else HealthStatus.CRITICAL
+                )
+
+            metrics.append(
+                HealthMetric(
+                    name="cpu_usage_percent",
+                    value=round(cpu_percent, 2),
+                    status=cpu_status,
+                    threshold=self.cpu_threshold,
+                    unit="%",
+                )
+            )
+        except Exception as e:
+            metrics.append(
+                HealthMetric(
+                    name="cpu_usage_percent",
+                    value=-1,
+                    status=HealthStatus.UNKNOWN,
+                    message=f"Failed to get CPU usage: {e}",
+                )
+            )
+        return metrics
+
+    def _check_memory_usage(self) -> List[HealthMetric]:
+        """Check memory usage."""
+        metrics = []
+        try:
+            memory_info = self.process.memory_info()
+            memory_mb = memory_info.rss / ResourceLimits.BYTES_TO_MB
+            memory_status = HealthStatus.HEALTHY
+            if memory_mb > self.memory_threshold_mb:
+                memory_status = (
+                    HealthStatus.WARNING
+                    if memory_mb < self.memory_threshold_mb * 1.2
+                    else HealthStatus.CRITICAL
+                )
+
+            metrics.append(
+                HealthMetric(
+                    name="memory_usage_mb",
+                    value=round(memory_mb, 2),
+                    status=memory_status,
+                    threshold=self.memory_threshold_mb,
+                    unit="MB",
+                )
+            )
+
+            metrics.append(
+                HealthMetric(
+                    name="memory_vms_mb",
+                    value=round(memory_info.vms / ResourceLimits.BYTES_TO_MB, 2),
+                    status=HealthStatus.HEALTHY,
+                    unit="MB",
+                )
+            )
+        except Exception as e:
+            metrics.append(
+                HealthMetric(
+                    name="memory_usage_mb",
+                    value=-1,
+                    status=HealthStatus.UNKNOWN,
+                    message=f"Failed to get memory usage: {e}",
+                )
+            )
+        return metrics
+
+    def _check_file_descriptors(self) -> List[HealthMetric]:
+        """Check file descriptor count (Unix only)."""
+        metrics = []
+        if hasattr(self.process, "num_fds"):
+            try:
+                fd_count = self.process.num_fds()
+                fd_status = HealthStatus.HEALTHY
+                if fd_count > self.fd_threshold:
+                    fd_status = (
+                        HealthStatus.WARNING
+                        if fd_count < self.fd_threshold * 1.2
+                        else HealthStatus.CRITICAL
+                    )
+
+                metrics.append(
+                    HealthMetric(
+                        name="file_descriptors",
+                        value=fd_count,
+                        status=fd_status,
+                        threshold=self.fd_threshold,
+                    )
+                )
+            except Exception as e:
+                metrics.append(
+                    HealthMetric(
+                        name="file_descriptors",
+                        value=-1,
+                        status=HealthStatus.UNKNOWN,
+                        message=f"Failed to get file descriptor count: {e}",
+                    )
+                )
+        return metrics
+
+    def _check_thread_count(self) -> List[HealthMetric]:
+        """Check thread count."""
+        metrics = []
+        try:
+            thread_count = self.process.num_threads()
+            metrics.append(
+                HealthMetric(
+                    name="thread_count",
+                    value=thread_count,
+                    status=HealthStatus.HEALTHY,
+                )
+            )
+        except Exception as e:
+            metrics.append(
+                HealthMetric(
+                    name="thread_count",
+                    value=-1,
+                    status=HealthStatus.UNKNOWN,
+                    message=f"Failed to get thread count: {e}",
+                )
+            )
+        return metrics
+
+    def _check_process_metadata(self) -> List[HealthMetric]:
+        """Check process metadata."""
+        metrics = []
+        try:
+            create_time = self.process.create_time()
+            metrics.append(
+                HealthMetric(
+                    name="process_start_time",
+                    value=create_time,
+                    status=HealthStatus.HEALTHY,
+                    unit="timestamp",
+                )
+            )
+        except Exception as e:
+            metrics.append(
+                HealthMetric(
+                    name="process_start_time",
+                    value=-1,
+                    status=HealthStatus.UNKNOWN,
+                    message=f"Failed to get process start time: {e}",
+                )
+            )
+        return metrics
+
+    def is_process_alive(self) -> bool:
+        """Quick check if process is still alive.
+
+        Returns:
+            True if process exists and is running
+        """
+        if not PSUTIL_AVAILABLE or not self.process:
+            return False
+
+        try:
+            return self.process.is_running()
+        except (psutil.NoSuchProcess, psutil.AccessDenied):
+            return False

claude_mpm/services/infrastructure/monitoring/resources.py

@@ -0,0 +1,243 @@
+"""Resource monitoring service for system resources (CPU, memory, disk).
+
+Monitors system-wide resource usage including CPU, memory, and disk utilization.
+"""
+
+from typing import Dict, List, Optional
+
+from .base import BaseMonitoringService, HealthMetric, HealthStatus
+
+try:
+    import psutil
+
+    PSUTIL_AVAILABLE = True
+except ImportError:
+    PSUTIL_AVAILABLE = False
+    psutil = None
+
+
+class ResourceMonitorService(BaseMonitoringService):
+    """Service for monitoring system resource usage.
+
+    Monitors:
+    - System CPU usage
+    - System memory usage
+    - Disk space utilization
+    - System load average
+    """
+
+    def __init__(
+        self,
+        cpu_threshold: float = 80.0,
+        memory_threshold: float = 85.0,
+        disk_threshold: float = 90.0,
+    ):
+        """Initialize resource monitor service.
+
+        Args:
+            cpu_threshold: CPU usage warning threshold (%)
+            memory_threshold: Memory usage warning threshold (%)
+            disk_threshold: Disk usage warning threshold (%)
+        """
+        super().__init__("ResourceMonitor")
+        self.cpu_threshold = cpu_threshold
+        self.memory_threshold = memory_threshold
+        self.disk_threshold = disk_threshold
+
+    async def check_health(self) -> List[HealthMetric]:
+        """Check system resource health."""
+        metrics = []
+
+        if not PSUTIL_AVAILABLE:
+            metrics.append(
+                HealthMetric(
+                    name="psutil_availability",
+                    value=False,
+                    status=HealthStatus.WARNING,
+                    message="psutil not available for resource monitoring",
+                )
+            )
+            return metrics
+
+        # CPU usage
+        try:
+            cpu_percent = psutil.cpu_percent(interval=0.1)
+            cpu_status = self._get_threshold_status(cpu_percent, self.cpu_threshold)
+
+            metrics.append(
+                HealthMetric(
+                    name="system_cpu_usage",
+                    value=round(cpu_percent, 2),
+                    status=cpu_status,
+                    threshold=self.cpu_threshold,
+                    unit="%",
+                )
+            )
+
+            # CPU count for context
+            metrics.append(
+                HealthMetric(
+                    name="cpu_count",
+                    value=psutil.cpu_count(),
+                    status=HealthStatus.HEALTHY,
+                )
+            )
+        except Exception as e:
+            self.logger.error(f"Failed to get CPU usage: {e}")
+            metrics.append(
+                HealthMetric(
+                    name="system_cpu_usage",
+                    value=-1,
+                    status=HealthStatus.UNKNOWN,
+                    message=str(e),
+                )
+            )
+
+        # Memory usage
+        try:
+            memory = psutil.virtual_memory()
+            memory_status = self._get_threshold_status(
+                memory.percent, self.memory_threshold
+            )
+
+            metrics.append(
+                HealthMetric(
+                    name="system_memory_usage",
+                    value=round(memory.percent, 2),
+                    status=memory_status,
+                    threshold=self.memory_threshold,
+                    unit="%",
+                )
+            )
+
+            # Memory details
+            metrics.append(
+                HealthMetric(
+                    name="memory_available_gb",
+                    value=round(memory.available / (1024**3), 2),
+                    status=HealthStatus.HEALTHY,
+                    unit="GB",
+                )
+            )
+
+            metrics.append(
+                HealthMetric(
+                    name="memory_total_gb",
+                    value=round(memory.total / (1024**3), 2),
+                    status=HealthStatus.HEALTHY,
+                    unit="GB",
+                )
+            )
+        except Exception as e:
+            self.logger.error(f"Failed to get memory usage: {e}")
+            metrics.append(
+                HealthMetric(
+                    name="system_memory_usage",
+                    value=-1,
+                    status=HealthStatus.UNKNOWN,
+                    message=str(e),
+                )
+            )
+
+        # Disk usage
+        try:
+            disk = psutil.disk_usage("/")
+            disk_status = self._get_threshold_status(disk.percent, self.disk_threshold)
+
+            metrics.append(
+                HealthMetric(
+                    name="disk_usage",
+                    value=round(disk.percent, 2),
+                    status=disk_status,
+                    threshold=self.disk_threshold,
+                    unit="%",
+                )
+            )
+
+            metrics.append(
+                HealthMetric(
+                    name="disk_free_gb",
+                    value=round(disk.free / (1024**3), 2),
+                    status=HealthStatus.HEALTHY,
+                    unit="GB",
+                )
+            )
+        except Exception as e:
+            self.logger.error(f"Failed to get disk usage: {e}")
+            metrics.append(
+                HealthMetric(
+                    name="disk_usage",
+                    value=-1,
+                    status=HealthStatus.UNKNOWN,
+                    message=str(e),
+                )
+            )
+
+        # Load average (Unix only)
+        try:
+            if hasattr(psutil, "getloadavg"):
+                load1, load5, load15 = psutil.getloadavg()
+                cpu_count = psutil.cpu_count()
+
+                # Load is concerning if > cpu_count
+                load_status = HealthStatus.HEALTHY
+                if load1 > cpu_count:
+                    load_status = HealthStatus.WARNING
+                if load1 > cpu_count * 1.5:
+                    load_status = HealthStatus.CRITICAL
+
+                metrics.append(
+                    HealthMetric(
+                        name="load_average_1min",
+                        value=round(load1, 2),
+                        status=load_status,
+                        threshold=cpu_count,
+                    )
+                )
+
+                metrics.append(
+                    HealthMetric(
+                        name="load_average_5min",
+                        value=round(load5, 2),
+                        status=HealthStatus.HEALTHY,
+                    )
+                )
+        except Exception as e:
+            self.logger.debug(f"Load average not available: {e}")
+
+        return metrics
+
+    def _get_threshold_status(self, value: float, threshold: float) -> HealthStatus:
+        """Determine health status based on threshold.
+
+        Args:
+            value: Current value
+            threshold: Warning threshold
+
+        Returns:
+            Health status based on value vs threshold
+        """
+        if value < threshold:
+            return HealthStatus.HEALTHY
+        if value < threshold * 1.1:  # 10% above threshold
+            return HealthStatus.WARNING
+        return HealthStatus.CRITICAL
+
+    def get_resource_summary(self) -> Optional[Dict[str, float]]:
+        """Get quick resource summary without full health check.
+
+        Returns:
+            Dictionary with current resource usage percentages
+        """
+        if not PSUTIL_AVAILABLE:
+            return None
+
+        try:
+            return {
+                "cpu_percent": psutil.cpu_percent(interval=0.1),
+                "memory_percent": psutil.virtual_memory().percent,
+                "disk_percent": psutil.disk_usage("/").percent,
+            }
+        except Exception as e:
+            self.logger.error(f"Failed to get resource summary: {e}")
+            return None