Add graceful degradation when external tools are missing

Checks availability of required (smartctl, lsblk) and optional (nvme,
ceph, pct, dmidecode) tools at startup. Guards all tool-dependent code
sections to skip gracefully with informative log messages instead of
crashing. Also fixes pre-existing indentation bug in LXC exception handler.

Fixes: #19

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

hwmonDaemon.py

@@ -1,5 +1,5 @@
 #!/usr/bin/env python3
-import os, sys, json, requests, psutil, socket, subprocess, logging, argparse, urllib.request, re, glob, datetime, fcntl, textwrap
+import os, sys, json, requests, psutil, socket, subprocess, logging, argparse, urllib.request, re, glob, datetime, fcntl, textwrap, shutil
 from typing import Dict, Any, List
 
 # =============================================================================
@@ -668,6 +668,41 @@ class SystemHealthMonitor:
         # Drive details cache (per-run, cleared on next execution)
         self._drive_details_cache = {}
 
+        # Check tool availability at startup
+        self._available_tools = self._check_tool_availability()
+
+    def _check_tool_availability(self) -> Dict[str, bool]:
+        """Check which external tools are available on this system.
+
+        Returns a dict mapping tool names to availability booleans.
+        Logs warnings for missing required tools and info for missing optional tools.
+        """
+        required_tools = {
+            'smartctl': 'smartmontools',
+            'lsblk': 'util-linux',
+        }
+        optional_tools = {
+            'nvme': 'nvme-cli',
+            'ceph': 'ceph-common',
+            'pct': 'pve-container',
+            'dmidecode': 'dmidecode',
+        }
+
+        availability = {}
+        for tool, package in required_tools.items():
+            available = shutil.which(tool) is not None
+            availability[tool] = available
+            if not available:
+                logger.warning(f"Required tool '{tool}' not found (install: apt install {package})")
+
+        for tool, package in optional_tools.items():
+            available = shutil.which(tool) is not None
+            availability[tool] = available
+            if not available:
+                logger.debug(f"Optional tool '{tool}' not found (install: apt install {package})")
+
+        return availability
+
     def _enforce_storage_limit(self, history_dir: str, max_bytes: int = None):
         """
         Delete oldest history files if directory exceeds size limit.
@@ -2588,8 +2623,8 @@ class SystemHealthMonitor:
             logger.debug(f"Detected Issues: {smart_health['issues']}")
             logger.debug("=== End SMART Check ===\n")
 
-            # Special handling for NVMe drives
+            # Special handling for NVMe drives (requires nvme-cli)
-            if 'nvme' in device:
+            if 'nvme' in device and self._available_tools.get('nvme'):
                 try:
                     nvme_result = subprocess.run(
                         ['nvme', 'smart-log', device],
@@ -2656,6 +2691,10 @@ class SystemHealthMonitor:
             'manufacturer_profile': None
         }
 
+        if not self._available_tools.get('nvme'):
+            logger.debug(f"nvme-cli not available, skipping NVMe health check for {device}")
+            return smart_health
+
         try:
             # Use nvme-cli for NVMe devices
             result = subprocess.run(
@@ -2722,6 +2761,11 @@ class SystemHealthMonitor:
         """Check health of all drives in the system."""
         drives_health = {'overall_status': 'NORMAL', 'drives': []}
 
+        if not self._available_tools.get('smartctl') or not self._available_tools.get('lsblk'):
+            logger.warning("Drive health checks skipped: smartctl or lsblk not available")
+            drives_health['overall_status'] = 'UNKNOWN'
+            return drives_health
+
         try:
             # Get only valid physical disks
             physical_disks = self._get_all_disks()
@@ -2846,7 +2890,8 @@ class SystemHealthMonitor:
         }
         
         try:
-            # First check using dmidecode
+            # First check using dmidecode (if available)
+            if self._available_tools.get('dmidecode'):
                 result = subprocess.run(
                     ['dmidecode', '--type', 'memory'],
                     stdout=subprocess.PIPE,
@@ -2857,7 +2902,7 @@ class SystemHealthMonitor:
                 if 'Error Correction Type: Multi-bit ECC' in result.stdout:
                     memory_health['has_ecc'] = True
 
-            # If dmidecode didn't find ECC, try the edac method as backup
+            # If dmidecode unavailable or didn't find ECC, try the edac method as backup
             if not memory_health['has_ecc']:
                 edac_path = '/sys/devices/system/edac/mc'
                 if os.path.exists(edac_path) and os.listdir(edac_path):
@@ -2994,8 +3039,6 @@ class SystemHealthMonitor:
         Returns health status, cluster info, and any issues detected.
         Cluster-wide issues use [cluster-wide] tag for cross-node deduplication.
         """
-        import shutil
-
         ceph_health = {
             'status': 'OK',
             'is_ceph_node': False,
@@ -3013,7 +3056,7 @@ class SystemHealthMonitor:
             return ceph_health
 
         # Check if ceph CLI is available
-        if not shutil.which('ceph'):
+        if not self._available_tools.get('ceph'):
             logger.debug("Ceph CLI not found - not a Ceph node")
             return ceph_health
 
@@ -3420,6 +3463,10 @@ class SystemHealthMonitor:
             'issues': []
         }
 
+        if not self._available_tools.get('pct'):
+            logger.debug("pct not available - not a PVE node or pve-container not installed")
+            return lxc_health
+
         try:
             result = subprocess.run(
                 ['pct', 'list'],