show more if fail refresh

src/fgai/monitor.py

@@ -1,6 +1,9 @@
 from __future__ import annotations
 
+import contextlib
 import json
+import signal
+import threading
 import time
 from pathlib import Path
 
@@ -28,6 +31,26 @@ from .triage import build_triage_queue
 from .stream_profiles import parse_profiles
 
 
+@contextlib.contextmanager
+def _cycle_timeout(seconds: int):
+    if seconds <= 0 or threading.current_thread() is not threading.main_thread() or not hasattr(signal, "SIGALRM"):
+        yield
+        return
+    previous_handler = signal.getsignal(signal.SIGALRM)
+    previous_timer = signal.getitimer(signal.ITIMER_REAL)
+
+    def _raise_timeout(_signum, _frame):
+        raise TimeoutError(f"status_cycle_timeout_{seconds}s")
+
+    signal.signal(signal.SIGALRM, _raise_timeout)
+    signal.setitimer(signal.ITIMER_REAL, seconds)
+    try:
+        yield
+    finally:
+        signal.setitimer(signal.ITIMER_REAL, previous_timer[0], previous_timer[1])
+        signal.signal(signal.SIGALRM, previous_handler)
+
+
 def _stream_titles(config: dict[str, object]) -> dict[str, str]:
     return {
         str(item.get("id", "")): str(item.get("title", "") or item.get("id", ""))
@@ -604,8 +627,8 @@ def monitor_loop(
     history_path: str | None = None,
     status_cache_path: str | None = None,
 ) -> None:
-    print(f"Monitoring {log_path}")
-    print(f"Writing status to {output}")
+    print(f"Monitoring {log_path}", flush=True)
+    print(f"Writing status to {output}", flush=True)
     last_llm_at = 0
     last_llm_text: str | None = None
     while True:
@@ -617,11 +640,13 @@ def monitor_loop(
         if runtime.get("log_source") == "graylog_mcp":
             write_refreshing_status(output, cache_path=status_cache_path, call_timeout_seconds=mcp_call_timeout, poll_timeout_seconds=mcp_poll_timeout, runtime_values=runtime)
         try:
-            status = build_status(
-                log_path, policy_path=policy_path, anomaly_limit=anomaly_limit,
-                baseline_path=baseline_path, config_path=config_path, history_path=history_path,
-                status_cache_path=status_cache_path,
-            )
+            status_timeout = mcp_poll_timeout + max(30, mcp_call_timeout * 2)
+            with _cycle_timeout(status_timeout if runtime.get("log_source") == "graylog_mcp" else 0):
+                status = build_status(
+                    log_path, policy_path=policy_path, anomaly_limit=anomaly_limit,
+                    baseline_path=baseline_path, config_path=config_path, history_path=history_path,
+                    status_cache_path=status_cache_path,
+                )
         except Exception as exc:
             error_status = {"status": "error", "error": f"monitor_error: {exc}", "call_timeout_seconds": mcp_call_timeout, "poll_timeout_seconds": mcp_poll_timeout}
             status = cached_status_with_error(status_cache_path, error_status) if status_cache_path else None