Merge PR #614: fix: resolve systemd restart loop with --replace flag

Authored by voidborne-d. Fixes #576.

Adds --replace flag to 'hermes gateway run' that terminates any existing
gateway instance (SIGTERM with SIGKILL fallback) before starting.
Updated systemd unit template with --replace, ExecStop, KillMode, and
TimeoutStopSec for robust service management.

gateway/run.py

@@ -2459,34 +2459,77 @@ def _start_cron_ticker(stop_event: threading.Event, adapters=None, interval: int
     logger.info("Cron ticker stopped")
 
 
-async def start_gateway(config: Optional[GatewayConfig] = None) -> bool:
+async def start_gateway(config: Optional[GatewayConfig] = None, replace: bool = False) -> bool:
     """
     Start the gateway and run until interrupted.
     
     This is the main entry point for running the gateway.
     Returns True if the gateway ran successfully, False if it failed to start.
     A False return causes a non-zero exit code so systemd can auto-restart.
+    
+    Args:
+        config: Optional gateway configuration override.
+        replace: If True, kill any existing gateway instance before starting.
+                 Useful for systemd services to avoid restart-loop deadlocks
+                 when the previous process hasn't fully exited yet.
     """
     # ── Duplicate-instance guard ──────────────────────────────────────
     # Prevent two gateways from running under the same HERMES_HOME.
     # The PID file is scoped to HERMES_HOME, so future multi-profile
     # setups (each profile using a distinct HERMES_HOME) will naturally
     # allow concurrent instances without tripping this guard.
-    from gateway.status import get_running_pid
+    import time as _time
+    from gateway.status import get_running_pid, remove_pid_file
     existing_pid = get_running_pid()
     if existing_pid is not None and existing_pid != os.getpid():
-        hermes_home = os.getenv("HERMES_HOME", "~/.hermes")
+        if replace:
-        logger.error(
+            logger.info(
-            "Another gateway instance is already running (PID %d, HERMES_HOME=%s). "
+                "Replacing existing gateway instance (PID %d) with --replace.",
-            "Use 'hermes gateway restart' to replace it, or 'hermes gateway stop' first.",
+                existing_pid,
-            existing_pid, hermes_home,
+            )
-        )
+            try:
-        print(
+                os.kill(existing_pid, signal.SIGTERM)
-            f"\n❌ Gateway already running (PID {existing_pid}).\n"
+            except ProcessLookupError:
-            f"   Use 'hermes gateway restart' to replace it,\n"
+                pass  # Already gone
-            f"   or 'hermes gateway stop' to kill it first.\n"
+            except PermissionError:
-        )
+                logger.error(
-        return False
+                    "Permission denied killing PID %d. Cannot replace.",
+                    existing_pid,
+                )
+                return False
+            # Wait up to 10 seconds for the old process to exit
+            for _ in range(20):
+                try:
+                    os.kill(existing_pid, 0)
+                    _time.sleep(0.5)
+                except (ProcessLookupError, PermissionError):
+                    break  # Process is gone
+            else:
+                # Still alive after 10s — force kill
+                logger.warning(
+                    "Old gateway (PID %d) did not exit after SIGTERM, sending SIGKILL.",
+                    existing_pid,
+                )
+                try:
+                    os.kill(existing_pid, signal.SIGKILL)
+                    _time.sleep(0.5)
+                except (ProcessLookupError, PermissionError):
+                    pass
+            remove_pid_file()
+        else:
+            hermes_home = os.getenv("HERMES_HOME", "~/.hermes")
+            logger.error(
+                "Another gateway instance is already running (PID %d, HERMES_HOME=%s). "
+                "Use 'hermes gateway restart' to replace it, or 'hermes gateway stop' first.",
+                existing_pid, hermes_home,
+            )
+            print(
+                f"\n❌ Gateway already running (PID {existing_pid}).\n"
+                f"   Use 'hermes gateway restart' to replace it,\n"
+                f"   or 'hermes gateway stop' to kill it first.\n"
+                f"   Or use 'hermes gateway run --replace' to auto-replace.\n"
+            )
+            return False
 
     # Sync bundled skills on gateway start (fast -- skips unchanged)
     try:

hermes_cli/gateway.py

@@ -154,19 +154,25 @@ def get_hermes_cli_path() -> str:
 # =============================================================================
 
 def generate_systemd_unit() -> str:
+    import shutil
     python_path = get_python_path()
     working_dir = str(PROJECT_ROOT)
     
+    hermes_cli = shutil.which("hermes") or f"{python_path} -m hermes_cli.main"
     return f"""[Unit]
 Description={SERVICE_DESCRIPTION}
 After=network.target
 
 [Service]
 Type=simple
-ExecStart={python_path} -m hermes_cli.main gateway run
+ExecStart={python_path} -m hermes_cli.main gateway run --replace
+ExecStop={hermes_cli} gateway stop
 WorkingDirectory={working_dir}
 Restart=on-failure
 RestartSec=10
+KillMode=mixed
+KillSignal=SIGTERM
+TimeoutStopSec=15
 StandardOutput=journal
 StandardError=journal
 
@@ -377,8 +383,15 @@ def launchd_status(deep: bool = False):
 # Gateway Runner
 # =============================================================================
 
-def run_gateway(verbose: bool = False):
+def run_gateway(verbose: bool = False, replace: bool = False):
-    """Run the gateway in foreground."""
+    """Run the gateway in foreground.
+    
+    Args:
+        verbose: Enable verbose logging output.
+        replace: If True, kill any existing gateway instance before starting.
+                 This prevents systemd restart loops when the old process
+                 hasn't fully exited yet.
+    """
     sys.path.insert(0, str(PROJECT_ROOT))
     
     from gateway.run import start_gateway
@@ -393,7 +406,7 @@ def run_gateway(verbose: bool = False):
     
     # Exit with code 1 if gateway fails to connect any platform,
     # so systemd Restart=on-failure will retry on transient errors
-    success = asyncio.run(start_gateway())
+    success = asyncio.run(start_gateway(replace=replace))
     if not success:
         sys.exit(1)
 
@@ -765,7 +778,8 @@ def gateway_command(args):
     # Default to run if no subcommand
     if subcmd is None or subcmd == "run":
         verbose = getattr(args, 'verbose', False)
-        run_gateway(verbose)
+        replace = getattr(args, 'replace', False)
+        run_gateway(verbose, replace=replace)
         return
 
     if subcmd == "setup":

hermes_cli/main.py

@@ -1315,6 +1315,8 @@ For more help on a command:
     # gateway run (default)
     gateway_run = gateway_subparsers.add_parser("run", help="Run gateway in foreground")
     gateway_run.add_argument("-v", "--verbose", action="store_true")
+    gateway_run.add_argument("--replace", action="store_true",
+                             help="Replace any existing gateway instance (useful for systemd)")
     
     # gateway start
     gateway_start = gateway_subparsers.add_parser("start", help="Start gateway service")