Add Voyager tunnel and monitoring telemetry

collectors/monitoring.py

@@ -0,0 +1,82 @@
+#!/usr/bin/env python3
+import re
+import sqlite3
+import subprocess
+
+CONTAINER = "uptime-kuma"
+DATABASE = "/opt/docker/uptime-kuma/data/kuma.db"
+
+
+def _failure(detail):
+    return {
+        "state": "offline", "health": "unavailable", "summary": "Unavailable",
+        "detail": detail, "container_available": False, "version": None,
+        "active_monitors": 0, "up_count": 0, "down_count": 0, "unknown_count": 0,
+    }
+
+
+def _container_info():
+    result = subprocess.run(
+        ["docker", "inspect", "--format",
+         "{{.State.Status}}\t{{.State.Health.Status}}\t{{.Config.Image}}",
+         CONTAINER], capture_output=True, text=True)
+    if result.returncode != 0:
+        return None
+    fields = result.stdout.strip().split("\t")
+    if len(fields) != 3:
+        return None
+    status, health, image = fields
+    match = re.search(r":([^:@]+)$", image)
+    return status, health, match.group(1) if match else None
+
+
+def _database_counts():
+    connection = sqlite3.connect(f"file:{DATABASE}?mode=ro", uri=True, timeout=5)
+    try:
+        rows = connection.execute(
+            """
+            SELECT m.id,
+                   (SELECT h.status FROM heartbeat h
+                    WHERE h.monitor_id = m.id
+                    ORDER BY h.id DESC LIMIT 1) AS latest_status
+            FROM monitor m
+            WHERE m.active = 1
+            """
+        ).fetchall()
+    finally:
+        connection.close()
+    counts = {"active": len(rows), "up": 0, "down": 0, "unknown": 0}
+    for _, status in rows:
+        if status == 1:
+            counts["up"] += 1
+        elif status == 0:
+            counts["down"] += 1
+        else:
+            counts["unknown"] += 1
+    return counts
+
+
+def collect():
+    container = _container_info()
+    if container is None:
+        return _failure("Uptime Kuma container unavailable")
+    container_status, container_health, version = container
+    if container_status != "running":
+        return _failure("Uptime Kuma container is not running")
+    try:
+        counts = _database_counts()
+    except (OSError, sqlite3.Error):
+        return _failure("Uptime Kuma aggregate query unavailable")
+    active, up = counts["active"], counts["up"]
+    down, unknown = counts["down"], counts["unknown"]
+    degraded = container_health == "unhealthy" or down > 0 or unknown > 0
+    version_detail = f"Uptime Kuma {version}" if version else "Uptime Kuma"
+    return {
+        "state": "warning" if degraded else "online",
+        "health": "partial" if degraded else "healthy",
+        "summary": f"{up}/{active} monitors up",
+        "detail": f"{version_detail} · edge01 monitoring operational",
+        "container_available": True, "container_health": container_health,
+        "version": version, "active_monitors": active, "up_count": up,
+        "down_count": down, "unknown_count": unknown,
+    }

collectors/wireguard.py

@@ -0,0 +1,63 @@
+#!/usr/bin/env python3
+import subprocess
+import time
+
+INTERFACE = "wg0"
+RECENT_HANDSHAKE_SECONDS = 30 * 60
+
+
+def _failure(detail):
+    return {
+        "state": "offline", "health": "unavailable", "summary": "Unavailable",
+        "detail": detail, "interface": INTERFACE, "interface_active": False,
+        "configured_peers": 0, "recent_handshakes": 0, "stale_handshakes": 0,
+        "rx_bytes": 0, "tx_bytes": 0,
+        "recent_handshake_threshold_seconds": RECENT_HANDSHAKE_SECONDS,
+    }
+
+
+def _format_bytes(value):
+    units = ("B", "KiB", "MiB", "GiB", "TiB")
+    amount = float(value)
+    for unit in units:
+        if amount < 1024 or unit == units[-1]:
+            return f"{amount:.2f} {unit}" if unit != "B" else f"{int(amount)} B"
+        amount /= 1024
+
+
+def collect():
+    result = subprocess.run(["wg", "show", INTERFACE, "dump"],
+                            capture_output=True, text=True)
+    if result.returncode != 0:
+        return _failure("WireGuard telemetry unavailable")
+    lines = [line.split() for line in result.stdout.splitlines() if line.strip()]
+    if not lines or len(lines[0]) < 4:
+        return _failure("WireGuard returned invalid telemetry")
+    peers = lines[1:]
+    now = int(time.time())
+    recent = rx_bytes = tx_bytes = 0
+    try:
+        for peer in peers:
+            if len(peer) < 8:
+                return _failure("WireGuard returned incomplete telemetry")
+            handshake = int(peer[4])
+            rx_bytes += int(peer[5])
+            tx_bytes += int(peer[6])
+            if handshake and now - handshake <= RECENT_HANDSHAKE_SECONDS:
+                recent += 1
+    except (TypeError, ValueError):
+        return _failure("WireGuard returned invalid telemetry")
+    configured = len(peers)
+    stale = configured - recent
+    state = "online" if configured and recent == configured else "warning"
+    health = "healthy" if state == "online" else "partial"
+    return {
+        "state": state, "health": health,
+        "summary": f"{recent}/{configured} peers recent",
+        "detail": (f"WireGuard active · {_format_bytes(rx_bytes)} RX · "
+                   f"{_format_bytes(tx_bytes)} TX"),
+        "interface": INTERFACE, "interface_active": True,
+        "configured_peers": configured, "recent_handshakes": recent,
+        "stale_handshakes": stale, "rx_bytes": rx_bytes, "tx_bytes": tx_bytes,
+        "recent_handshake_threshold_seconds": RECENT_HANDSHAKE_SECONDS,
+    }