Repair Local Agent supervisor and add update progress UI

app/modules/documents/local_agent_runtime/ERPAgentDashboard.pyw

@@ -1,14 +1,12 @@
 from __future__ import annotations
 
+import atexit
 import ctypes
 from ctypes import wintypes
 import hashlib
 import os
-import atexit
 from pathlib import Path
 import subprocess
-import sys
-import threading
 import time
 import urllib.request
 
@@ -16,8 +14,6 @@ from tkinter import Tk, messagebox
 
 
 INSTALL_ROOT = Path(__file__).resolve().parent
-DATA_DIR = INSTALL_ROOT / "data"
-SUPERVISOR_LOCK = DATA_DIR / "supervisor.lock"
 SUPERVISOR_SCRIPT = INSTALL_ROOT / "ERPAgentSupervisor.pyw"
 URL = "http://127.0.0.1:8788"
 ERROR_ALREADY_EXISTS = 183
@@ -69,7 +65,7 @@ def _focus_existing_dashboard() -> None:
             buf = ctypes.create_unicode_buffer(length + 1)
             user32.GetWindowTextW(hwnd, buf, length + 1)
             if "ERP Local Agent Dashboard" in buf.value:
-                user32.ShowWindow(hwnd, 9)  # SW_RESTORE
+                user32.ShowWindow(hwnd, 9)
                 user32.SetForegroundWindow(hwnd)
                 return False
             return True
@@ -93,42 +89,6 @@ def _popup(title: str, message: str, error: bool = False) -> None:
     root.destroy()
 
 
-def _pid_alive(pid: int) -> bool:
-    if pid <= 0:
-        return False
-    if os.name == "nt":
-        try:
-            import ctypes
-            PROCESS_QUERY_LIMITED_INFORMATION = 0x1000
-            STILL_ACTIVE = 259
-            kernel32 = ctypes.windll.kernel32
-            handle = kernel32.OpenProcess(PROCESS_QUERY_LIMITED_INFORMATION, False, int(pid))
-            if not handle:
-                return False
-            try:
-                exit_code = ctypes.c_ulong()
-                if not kernel32.GetExitCodeProcess(handle, ctypes.byref(exit_code)):
-                    return False
-                return exit_code.value == STILL_ACTIVE
-            finally:
-                kernel32.CloseHandle(handle)
-        except Exception:
-            return False
-    try:
-        os.kill(pid, 0)
-        return True
-    except OSError:
-        return False
-
-def _supervisor_running() -> bool:
-    try:
-        if not SUPERVISOR_LOCK.exists():
-            return False
-        return _pid_alive(int(SUPERVISOR_LOCK.read_text(encoding="utf-8").strip()))
-    except Exception:
-        return False
-
-
 def _pythonw() -> Path:
     pyw = INSTALL_ROOT / ".venv" / "Scripts" / "pythonw.exe"
     if pyw.exists():
@@ -139,18 +99,30 @@ def _pythonw() -> Path:
     raise RuntimeError("ERP Local Agent Python environment was not found.")
 
 
+def _ready(timeout: float = 1.5) -> bool:
+    try:
+        with urllib.request.urlopen(URL + "/api/status", timeout=timeout) as response:
+            return int(getattr(response, "status", 200)) < 500
+    except Exception:
+        return False
+
+
 def _ensure_supervisor() -> None:
-    if _supervisor_running():
+    # If the dashboard is already healthy, never start another supervisor or worker.
+    if _ready():
         return
 
-    # The Scheduled Task is the canonical owner/launcher. Starting it repeatedly
-    # is safe because the task is configured with MultipleInstances=IgnoreNew,
-    # and the supervisor itself also owns a machine-wide Global mutex.
     started_by_task = False
     if os.name == "nt":
         result = subprocess.run(
-            ["powershell.exe", "-NoProfile", "-ExecutionPolicy", "Bypass", "-Command",
-             "Start-ScheduledTask -TaskName 'ERP Local Agent' -ErrorAction Stop"],
+            [
+                "powershell.exe",
+                "-NoProfile",
+                "-ExecutionPolicy",
+                "Bypass",
+                "-Command",
+                "Start-ScheduledTask -TaskName 'ERP Local Agent' -ErrorAction Stop",
+            ],
             cwd=str(INSTALL_ROOT),
             stdout=subprocess.DEVNULL,
             stderr=subprocess.DEVNULL,
@@ -160,8 +132,7 @@ def _ensure_supervisor() -> None:
         started_by_task = result.returncode == 0
 
     if not started_by_task:
-        # Recovery fallback for a workstation where the scheduled task was not
-        # installed yet. The Global mutex still prevents a second supervisor.
+        # Recovery fallback. The Supervisor Global mutex makes repeated launches harmless.
         subprocess.Popen(
             [str(_pythonw()), str(SUPERVISOR_SCRIPT), "--background"],
             cwd=str(INSTALL_ROOT),
@@ -171,23 +142,8 @@ def _ensure_supervisor() -> None:
             close_fds=True,
         )
 
-    deadline = time.time() + 20
-    while time.time() < deadline:
-        if _supervisor_running():
-            return
-        time.sleep(0.5)
-    raise RuntimeError("ERP Local Agent supervisor did not start.")
 
-
-def _ready(timeout: float = 1.5) -> bool:
-    try:
-        with urllib.request.urlopen(URL + "/api/status", timeout=timeout) as response:
-            return int(getattr(response, "status", 200)) < 500
-    except Exception:
-        return False
-
-
-def _wait_ready(seconds: int = 75) -> bool:
+def _wait_ready(seconds: int = 90) -> bool:
     deadline = time.time() + seconds
     while time.time() < deadline:
         if _ready():
@@ -202,6 +158,7 @@ def main() -> None:
         _focus_existing_dashboard()
         return
     atexit.register(_close_dashboard_mutex, dashboard_mutex)
+
     try:
         _ensure_supervisor()
     except Exception as exc:
@@ -211,8 +168,9 @@ def main() -> None:
     if not _wait_ready():
         _popup(
             "ERP Local Agent",
-            "The Local Agent supervisor is running, but the dashboard did not become ready.\n\n"
-            f"Check logs under:\n{INSTALL_ROOT / 'logs'}",
+            "The Local Agent did not become ready.\n\n"
+            "Check the 'ERP Local Agent' Scheduled Task and the logs under:\n"
+            f"{INSTALL_ROOT / 'logs'}",
             error=True,
         )
         return
@@ -252,9 +210,9 @@ def main() -> None:
                         pass
                     outage_seen = False
                 was_ready = current
-                time.sleep(1.5)
+                time.sleep(1.0)
             except Exception:
-                time.sleep(2)
+                time.sleep(1.5)
 
     webview.start(monitor)