jean-router: ajout Wake-on-LAN

- Detection si Jean/Ollama inaccessible
- Envoi magic packet WoL vers TARGET_MAC (fallback PC_GAMING_MAC)
- Attente WAKE_TIMEOUT secondes que la machine demarre
- Integration dans get_presets, get_ollama_models, chat_completions
- Verrous pour eviter les WoL simultanes
- Active via ENABLE_WOL=true

jean-router/app.py

@@ -9,6 +9,7 @@ import os
 import time
 import logging
 import threading
+import socket
 from typing import Optional
 
 import requests
@@ -30,6 +31,12 @@ SWITCH_TIMEOUT = int(os.getenv("SWITCH_TIMEOUT", "90"))
 CHECK_INTERVAL = int(os.getenv("CHECK_INTERVAL", "2"))
 JEAN_STOP_TIMEOUT = int(os.getenv("JEAN_STOP_TIMEOUT", "30"))
 OLLAMA_UNLOAD_TIMEOUT = int(os.getenv("OLLAMA_UNLOAD_TIMEOUT", "15"))
+ENABLE_WOL = os.getenv("ENABLE_WOL", "false").lower() == "true"
+TARGET_MAC = os.getenv("TARGET_MAC", os.getenv("PC_GAMING_MAC", ""))
+BROADCAST_IP = os.getenv("BROADCAST_IP", "192.168.1.255")
+WAKE_TIMEOUT = int(os.getenv("WAKE_TIMEOUT", "120"))
+wol_in_progress = threading.Lock()
+jean_alive_cache = {"alive": False, "timestamp": 0}
 
 logging.basicConfig(level=logging.INFO, format="%(asctime)s [%(levelname)s] %(message)s")
 log = logging.getLogger(__name__)
@@ -65,10 +72,83 @@ def get_presets(force_refresh=False):
             return presets_cache["data"]
     except Exception as e:
         log.warning(f"[Presets] Erreur récupération: {e}")
+        if ENABLE_WOL and (not presets_cache["data"] or force_refresh):
+            wait_for_jean(trigger_wol=True)
+            try:
+                r = requests.get(f"http://{JEAN_WEB_HOST}:{JEAN_WEB_PORT}/api/presets", timeout=5)
+                if r.ok:
+                    presets_cache["data"] = r.json()
+                    presets_cache["timestamp"] = now
+                    return presets_cache["data"]
+            except Exception:
+                pass
 
     return presets_cache["data"]
 
 
+def send_wol():
+    """Envoie un magic packet WoL vers TARGET_MAC"""
+    if not TARGET_MAC:
+        log.warning("[WoL] TARGET_MAC non defini, WoL desactive")
+        return
+
+    mac = TARGET_MAC.replace(":", "").replace("-", "")
+    if len(mac) != 12:
+        log.error(f"[WoL] MAC invalide: {TARGET_MAC}")
+        return
+
+    data = bytes.fromhex("FF" * 6 + mac * 16)
+    with socket.socket(socket.AF_INET, socket.SOCK_DGRAM) as s:
+        s.setsockopt(socket.SOL_SOCKET, socket.SO_BROADCAST, 1)
+        for i in range(3):
+            s.sendto(data, (BROADCAST_IP, 9))
+            log.info(f"[WoL] Magic packet #{i+1} envoye vers {TARGET_MAC}")
+
+
+def jean_alive() -> bool:
+    """Verifie si Jean est accessible"""
+    now = time.time()
+    if now - jean_alive_cache["timestamp"] < 5:
+        return jean_alive_cache["alive"]
+
+    try:
+        url = f"http://{JEAN_WEB_HOST}:{JEAN_WEB_PORT}/api/presets"
+        r = requests.get(url, timeout=3)
+        alive = r.ok
+    except Exception:
+        alive = False
+
+    jean_alive_cache["alive"] = alive
+    jean_alive_cache["timestamp"] = now
+    return alive
+
+
+def wait_for_jean(trigger_wol=True):
+    """Attend que Jean soit accessible. Si ENABLE_WOL et trigger_wol, envoie WoL d'abord."""
+    if jean_alive():
+        return
+
+    if not ENABLE_WOL or not trigger_wol:
+        return
+
+    with wol_in_progress:
+        if jean_alive():
+            return
+
+        log.info("[WoL] Jean inaccessible, envoi WoL...")
+        send_wol()
+
+        deadline = time.time() + WAKE_TIMEOUT
+        while time.time() < deadline:
+            if jean_alive():
+                jean_alive_cache["timestamp"] = 0
+                log.info(f"[WoL] Jean UP apres {int(time.time() - deadline + WAKE_TIMEOUT)}s")
+                return
+            time.sleep(CHECK_INTERVAL)
+
+        log.error(f"[WoL] Jean ne repond pas apres {WAKE_TIMEOUT}s")
+
+
 def get_ollama_models(force_refresh=False):
     """Récupère la liste des modèles depuis Ollama"""
     if not ENABLE_OLLAMA:
@@ -88,6 +168,17 @@ def get_ollama_models(force_refresh=False):
             return data
     except Exception as e:
         log.warning(f"[Ollama] Erreur récupération: {e}")
+        if ENABLE_WOL and (not ollama_cache["data"] or force_refresh):
+            wait_for_jean(trigger_wol=True)
+            try:
+                r = requests.get(f"http://{OLLAMA_PROXY_HOST}:{OLLAMA_PROXY_PORT}/api/tags", timeout=5)
+                if r.ok:
+                    data = r.json().get("models", [])
+                    ollama_cache["data"] = data
+                    ollama_cache["timestamp"] = now
+                    return data
+            except Exception:
+                pass
 
     return ollama_cache["data"]
 
@@ -667,6 +758,8 @@ def chat_completions():
     if model_id in jean_preset_ids or model_id in jean_names:
         internal_id = jean_name_to_id.get(model_id, model_id)
 
+        wait_for_jean(trigger_wol=True)
+
         if not ensure_vram_for_backend("jean"):
             return jsonify({"error": "Failed to prepare VRAM for Jean"}), 503
 
@@ -695,6 +788,7 @@ def chat_completions():
         upstream_url = f"http://{JEAN_PROXY_HOST}:{JEAN_PROXY_PORT}/v1/chat/completions"
         log.info(f"[Router] → Jean ({model_id})")
     elif ENABLE_OLLAMA:
+        wait_for_jean(trigger_wol=True)
         ollama_models = get_ollama_models()
         ollama_ids = [m.get("name", "").replace(":latest", "") for m in ollama_models]