auto-sync: 2026-04-19 18:30:01

tasks/flightradar24/backup/backup.sh

@@ -0,0 +1,24 @@
+#!/bin/bash
+set -euo pipefail
+
+BACKUP_DIR="/backup"
+TIMESTAMP=$(date +%Y%m%d_%H%M%S)
+DUMP_FILE="${BACKUP_DIR}/fr24_${TIMESTAMP}.sql.gz"
+KEEP_LAST=7
+
+echo "[backup] starting dump at ${TIMESTAMP}"
+
+pg_dump \
+  -h "${PGHOST:-postgres}" \
+  -U "${PGUSER:-fr24}" \
+  -d "${PGDATABASE:-fr24}" \
+  --no-password \
+  | gzip > "${DUMP_FILE}"
+
+echo "[backup] dump written: ${DUMP_FILE} ($(du -sh "${DUMP_FILE}" | cut -f1))"
+
+# Keep only last N dumps
+cd "${BACKUP_DIR}"
+ls -1t fr24_*.sql.gz 2>/dev/null | tail -n +$((KEEP_LAST + 1)) | xargs -r rm -v
+
+echo "[backup] done. kept last ${KEEP_LAST} dumps."

tasks/flightradar24/compose/docker-compose.yml

@@ -129,19 +129,17 @@ services:
       - fr24-net
 
   monitoring:
-    image: alpine:3.20
+    build:
+      context: ../monitoring
+      dockerfile: Dockerfile
+    image: fr24-monitoring
     container_name: fr24-monitoring
-    command: ["sh", "-c", "echo 'monitoring placeholder: disk, db, capture status, alerts'; tail -f /dev/null"]
     environment:
       <<: *common-env
       SERVICE_ROLE: monitoring
-      MONITORING_INTERVAL_SECONDS: ${MONITORING_INTERVAL_SECONDS:-30}
-    ports:
-      - "${MONITORING_PUBLISHED_PORT:-9090}:9090"
+      MONITORING_INTERVAL_SECONDS: ${MONITORING_INTERVAL_SECONDS:-60}
     volumes:
       - ../logs/monitoring:/var/log/fr24
-      - ../backup:/backup
-      - /var/run/docker.sock:/var/run/docker.sock:ro
     depends_on:
       postgres:
         condition: service_healthy
@@ -152,11 +150,36 @@ services:
       api:
         condition: service_started
     healthcheck:
-      test: ["CMD-SHELL", "test -f /tmp/monitoring-ready || exit 1"]
+      test: ["CMD-SHELL", "test -f /tmp/monitoring-ready && pgrep -f main.py > /dev/null || exit 1"]
       interval: 30s
       timeout: 5s
       retries: 3
-      start_period: 10s
+      start_period: 15s
+    restart: unless-stopped
+    networks:
+      - fr24-net
+
+  backup:
+    image: postgres:16-alpine
+    container_name: fr24-backup
+    environment:
+      PGPASSWORD: ${POSTGRES_PASSWORD:-change-me}
+      PGHOST: ${POSTGRES_HOST:-postgres}
+      PGUSER: ${POSTGRES_USER:-fr24}
+      PGDATABASE: ${POSTGRES_DB:-fr24}
+      TZ: ${TZ:-UTC}
+    volumes:
+      - ../backup:/backup
+      - ../backup/backup.sh:/usr/local/bin/backup.sh:ro
+    depends_on:
+      postgres:
+        condition: service_healthy
+    command: >
+      sh -c 'echo "[backup] service started, interval=6h" &&
+             while true; do
+               /usr/local/bin/backup.sh;
+               sleep 21600;
+             done'
     restart: unless-stopped
     networks:
       - fr24-net

tasks/flightradar24/monitoring/Dockerfile

@@ -0,0 +1,13 @@
+FROM python:3.11-slim
+
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    libpq-dev gcc python3-dev \
+    && rm -rf /var/lib/apt/lists/*
+
+WORKDIR /app
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+COPY main.py .
+
+CMD ["python", "-u", "main.py"]

tasks/flightradar24/monitoring/main.py

@@ -0,0 +1,114 @@
+#!/usr/bin/env python3
+"""FR24 monitoring service — checks disk, DB size, capture lag, throughput every 60s."""
+
+import os
+import time
+import logging
+import subprocess
+import psycopg2
+
+logging.basicConfig(
+    level=logging.INFO,
+    format="%(asctime)s %(levelname)s %(message)s",
+    datefmt="%Y-%m-%dT%H:%M:%S",
+)
+log = logging.getLogger("monitor")
+
+DB_DSN = (
+    f"host={os.environ.get('POSTGRES_HOST', 'postgres')} "
+    f"port={os.environ.get('POSTGRES_PORT', '5432')} "
+    f"dbname={os.environ.get('POSTGRES_DB', 'fr24')} "
+    f"user={os.environ.get('POSTGRES_USER', 'fr24')} "
+    f"password={os.environ.get('POSTGRES_PASSWORD', 'change-me')}"
+)
+INTERVAL = int(os.environ.get("MONITORING_INTERVAL_SECONDS", "60"))
+DISK_WARN_PCT = 80
+LAG_WARN_SEC = 300  # 5 minutes
+
+
+def get_disk_usage() -> str:
+    """Return disk usage percent for / as integer string, e.g. '45'."""
+    try:
+        result = subprocess.run(
+            ["df", "-P", "/"],
+            capture_output=True, text=True, timeout=5
+        )
+        # last line: Filesystem 1024-blocks Used Available Capacity Mounted
+        line = result.stdout.strip().splitlines()[-1]
+        pct = line.split()[4].rstrip("%")
+        return pct
+    except Exception as e:
+        log.warning("disk check failed: %s", e)
+        return "?"
+
+
+def run_checks():
+    disk_pct_str = get_disk_usage()
+
+    try:
+        conn = psycopg2.connect(DB_DSN, connect_timeout=5)
+        conn.autocommit = True
+        cur = conn.cursor()
+
+        # DB size
+        cur.execute("SELECT pg_database_size(current_database())")
+        db_bytes = cur.fetchone()[0]
+        db_size_gb = db_bytes / (1024 ** 3)
+        db_size_str = f"{db_size_gb:.2f}GB" if db_size_gb >= 1 else f"{db_bytes / (1024**2):.1f}MB"
+
+        # Capture lag
+        cur.execute("SELECT EXTRACT(EPOCH FROM (now() - MAX(observed_at))) FROM fr24.raw_packets")
+        row = cur.fetchone()
+        lag_sec = int(row[0]) if row and row[0] is not None else None
+        lag_str = f"{lag_sec}s" if lag_sec is not None else "N/A"
+
+        # Throughput: packets in last 5 minutes
+        cur.execute(
+            "SELECT COUNT(*) FROM fr24.raw_packets "
+            "WHERE observed_at >= now() - INTERVAL '5 minutes'"
+        )
+        throughput = cur.fetchone()[0]
+
+        cur.close()
+        conn.close()
+
+        db_ok = True
+    except Exception as e:
+        log.warning("db check failed: %s", e)
+        db_size_str = "ERR"
+        lag_sec = None
+        lag_str = "ERR"
+        throughput = "ERR"
+        db_ok = False
+
+    # Emit metrics line
+    disk_display = f"{disk_pct_str}%" if disk_pct_str != "?" else "?"
+    print(
+        f"[monitor] disk={disk_display} db_size={db_size_str} "
+        f"capture_lag={lag_str} throughput={throughput}pkt/5min",
+        flush=True,
+    )
+
+    # Warnings
+    if disk_pct_str not in ("?",) and int(disk_pct_str) > DISK_WARN_PCT:
+        log.warning("DISK USAGE HIGH: %s%%", disk_pct_str)
+
+    if db_ok and lag_sec is not None and lag_sec > LAG_WARN_SEC:
+        log.warning("CAPTURE LAG HIGH: %ds (threshold %ds)", lag_sec, LAG_WARN_SEC)
+
+
+def main():
+    log.info("FR24 monitoring started (interval=%ds)", INTERVAL)
+    # Signal readiness
+    open("/tmp/monitoring-ready", "w").close()
+
+    while True:
+        try:
+            run_checks()
+        except Exception as e:
+            log.error("unexpected error in run_checks: %s", e)
+        time.sleep(INTERVAL)
+
+
+if __name__ == "__main__":
+    main()

tasks/flightradar24/monitoring/requirements.txt

@@ -0,0 +1 @@
+psycopg2-binary==2.9.9