fix: restore self-healing runtime checks

scripts/self_healing.py

@@ -48,6 +48,34 @@ class SelfHealer:
             self.log(f"  [ERROR] Failed to run remote command on {host}: {e}")
             return None
 
+    def confirm(self, prompt: str) -> bool:
+        """Ask for confirmation unless --yes flag is set."""
+        if self.yes:
+            return True
+        while True:
+            response = input(f"{prompt} [y/N] ").strip().lower()
+            if response in ("y", "yes"):
+                return True
+            if response in ("n", "no", ""):
+                return False
+            print("Please answer 'y' or 'n'.")
+
+    def check_llama_server(self, host: str):
+        ip = FLEET[host]["ip"]
+        port = FLEET[host]["port"]
+        try:
+            requests.get(f"http://{ip}:{port}/health", timeout=2)
+        except requests.RequestException:
+            self.log(f"  [!] llama-server down on {host}.")
+            if self.dry_run:
+                self.log(f"  [DRY-RUN] Would restart llama-server on {host}")
+            else:
+                if self.confirm(f"  Restart llama-server on {host}?"):
+                    self.log(f"  Restarting llama-server on {host}...")
+                    self.run_remote(host, "systemctl restart llama-server")
+                else:
+                    self.log(f"  Skipped restart on {host}.")
+
     def check_disk_space(self, host: str):
         res = self.run_remote(host, "df -h / | tail -1 | awk '{print $5}' | sed 's/%//'")
         if res and res.returncode == 0:

tests/test_self_healing.py

@@ -3,6 +3,7 @@
 from __future__ import annotations
 
 import importlib.util
+import subprocess
 from pathlib import Path
 from unittest.mock import MagicMock
 
@@ -60,3 +61,35 @@ class TestMainCli:
 
         healer_cls.assert_called_once_with(dry_run=False, confirm_kill=True, yes=True)
         healer.run.assert_called_once_with()
+
+    def test_real_default_dry_run_path_completes(self, monkeypatch, capsys):
+        class FakeExecutor:
+            def __init__(self):
+                self.calls = []
+
+            def run_script(self, host, command, *, local=False, timeout=None):
+                self.calls.append((host, command, local, timeout))
+                if command.startswith("df -h /"):
+                    return subprocess.CompletedProcess(command, 0, stdout="42\n", stderr="")
+                if command.startswith("free -m"):
+                    return subprocess.CompletedProcess(command, 0, stdout="12.5\n", stderr="")
+                if command.startswith("ps aux"):
+                    return subprocess.CompletedProcess(command, 0, stdout="", stderr="")
+                raise AssertionError(f"unexpected command: {command}")
+
+        fake_executor = FakeExecutor()
+        monkeypatch.setattr(sh, "FLEET", {"mac": {"ip": "127.0.0.1", "port": 8080}})
+        monkeypatch.setattr(sh.requests, "get", lambda url, timeout: object())
+        monkeypatch.setattr(sh, "VerifiedSSHExecutor", lambda: fake_executor)
+
+        sh.main([])
+
+        out = capsys.readouterr().out
+        assert "Starting self-healing cycle (DRY-RUN mode)." in out
+        assert "Auditing mac..." in out
+        assert "Cycle complete." in out
+        assert fake_executor.calls == [
+            ("127.0.0.1", "df -h / | tail -1 | awk '{print $5}' | sed 's/%//'", True, 15),
+            ("127.0.0.1", "free -m | awk '/^Mem:/{print $3/$2 * 100}'", True, 15),
+            ("127.0.0.1", "ps aux --sort=-%cpu | awk 'NR>1 && $3>80 {print $2, $11, $3}'", True, 15),
+        ]