fix request handling

jtl-wafi-agent.py

@@ -1331,9 +1331,9 @@ class LiveStatsTracker:
         self.blocked_paths: Dict[str, List[float]] = {}  # Path -> [blocked timestamps]
         self.ip_404s: Dict[str, List[float]] = {}  # IP -> [404 timestamps]
 
-        # IP Request Details (für IP-Detail Popup)
+        # IP Request Details (für IP-Detail Popup) - werden NICHT nach window_seconds gelöscht!
         self.ip_request_details: Dict[str, List[Dict]] = {}  # IP -> [{ts, path, ua}, ...]
-        self.max_request_details = 100  # Max Anzahl Details pro IP
+        self.max_request_details = 500  # Max Anzahl Details pro IP (nicht zeitbasiert)
 
         # Human/Bot Counters
         self.human_requests: List[float] = []
@@ -1357,15 +1357,12 @@ class LiveStatsTracker:
         self.human_requests = [t for t in self.human_requests if t > cutoff]
         self.bot_requests = [t for t in self.bot_requests if t > cutoff]
 
-        # Request Details cleanup
+        # Request Details werden NICHT zeitbasiert gelöscht - nur durch max_request_details limitiert
-        for ip in list(self.ip_request_details.keys()):
+        # Damit bleiben alle Requests im Speicher für das IP-Detail-Popup
-            self.ip_request_details[ip] = [r for r in self.ip_request_details[ip] if r['ts'] > cutoff]
-            if not self.ip_request_details[ip]:
-                del self.ip_request_details[ip]
 
-    def record_request(self, ip: str, path: str, is_bot: bool, is_blocked: bool = False, is_404: bool = False, user_agent: str = ''):
+    def record_request(self, ip: str, path: str, is_bot: bool, is_blocked: bool = False, is_404: bool = False, user_agent: str = '', timestamp: float = None):
-        """Zeichnet einen Request auf."""
+        """Zeichnet einen Request auf. timestamp=None nutzt aktuelle Zeit."""
-        now = time.time()
+        now = timestamp if timestamp is not None else time.time()
 
         # IP-Request
         if ip not in self.ip_requests:
@@ -1526,13 +1523,102 @@ _shop_stats_trackers: Dict[str, LiveStatsTracker] = {}
 def get_shop_stats_tracker(shop: str, window_seconds: int = 300) -> LiveStatsTracker:
     """Gibt den Stats-Tracker für einen Shop zurück (erstellt falls nötig)."""
     if shop not in _shop_stats_trackers:
-        _shop_stats_trackers[shop] = LiveStatsTracker(shop, window_seconds)
+        tracker = LiveStatsTracker(shop, window_seconds)
+        # Initialen Log-Scan durchführen um Request-Details zu laden
+        _load_initial_request_details(tracker, shop)
+        _shop_stats_trackers[shop] = tracker
     elif _shop_stats_trackers[shop].window_seconds != window_seconds:
         # Window geändert - neuen Tracker erstellen
-        _shop_stats_trackers[shop] = LiveStatsTracker(shop, window_seconds)
+        tracker = LiveStatsTracker(shop, window_seconds)
+        _load_initial_request_details(tracker, shop)
+        _shop_stats_trackers[shop] = tracker
     return _shop_stats_trackers[shop]
 
 
+def _load_initial_request_details(tracker: LiveStatsTracker, shop: str, max_lines: int = 5000):
+    """
+    Lädt initiale Request-Details aus dem Log-File.
+    Wird beim Erstellen eines neuen LiveStatsTracker aufgerufen.
+    """
+    log_file = os.path.join(VHOSTS_DIR, shop, 'httpdocs', SHOP_LOG_FILE)
+
+    if not os.path.isfile(log_file):
+        return
+
+    try:
+        # Letzte N Zeilen des Logs lesen (effizient von hinten)
+        with open(log_file, 'r') as f:
+            # Gehe ans Ende der Datei
+            f.seek(0, 2)
+            file_size = f.tell()
+
+            if file_size == 0:
+                return
+
+            # Lies maximal 500KB von hinten
+            read_size = min(512000, file_size)
+            f.seek(file_size - read_size)
+
+            # Überspringe erste (möglicherweise unvollständige) Zeile
+            if read_size < file_size:
+                f.readline()
+
+            lines = f.readlines()
+
+        # Nur die letzten max_lines Zeilen verarbeiten
+        lines = lines[-max_lines:]
+
+        for line in lines:
+            try:
+                # IP extrahieren
+                ip = None
+                if 'IP: ' in line:
+                    ip = line.split('IP: ')[1].split(' |')[0].strip()
+
+                if not ip:
+                    continue
+
+                # Zeitstempel extrahieren (Format: [2024-01-09 12:34:56])
+                timestamp = None
+                if line.startswith('[') and ']' in line:
+                    try:
+                        ts_str = line[1:line.index(']')]
+                        ts_dt = datetime.strptime(ts_str, '%Y-%m-%d %H:%M:%S')
+                        timestamp = ts_dt.timestamp()
+                    except:
+                        pass
+
+                # Path extrahieren
+                path = '/'
+                if 'Path: ' in line:
+                    path = line.split('Path: ')[1].split(' |')[0].strip()
+                elif 'URI: ' in line:
+                    path = line.split('URI: ')[1].split(' |')[0].strip()
+
+                # User-Agent extrahieren
+                user_agent = ''
+                if 'UA: ' in line:
+                    user_agent = line.split('UA: ')[1].split(' |')[0].strip()
+                elif 'User-Agent: ' in line:
+                    user_agent = line.split('User-Agent: ')[1].split(' |')[0].strip()
+
+                # Bot/Blocked erkennen
+                is_bot = any(x in line for x in ['BOT: ', 'BOT:', 'BLOCKED_BOT:', 'MONITOR_BOT:', 'BANNED_BOT:'])
+                is_blocked = any(x in line for x in ['BANNED', 'BLOCKED'])
+                is_404 = '404' in line
+
+                # Request aufzeichnen (mit originalem Zeitstempel falls vorhanden)
+                tracker.record_request(ip, path, is_bot, is_blocked, is_404, user_agent, timestamp)
+
+            except Exception:
+                continue
+
+        logger.info(f"Initiale Request-Details für {shop} geladen: {len(tracker.ip_request_details)} IPs")
+
+    except Exception as e:
+        logger.warning(f"Fehler beim Laden der initialen Request-Details für {shop}: {e}")
+
+
 # =============================================================================
 # BAN/WHITELIST MANAGEMENT
 # =============================================================================