Add ping check and latency display

DOCKER_README.md

@@ -4,7 +4,7 @@
 ![Docker Image Size](https://img.shields.io/docker/image-size/syntaxsorcerer7/internet-monitor)
 ![Docker Stars](https://img.shields.io/docker/stars/syntaxsorcerer7/internet-monitor)
 
-A containerized internet monitoring tool that continuously tracks your internet connection availability and visualizes it in a clear web interface with professional charts.
+A containerized internet monitoring tool that continuously tracks your internet connection availability and ping latency and visualizes it in a clear web interface with professional charts.
 
 ## 🚀 Quick Start
 
@@ -22,9 +22,9 @@ open http://localhost:8000
 ## 📊 Features
 
 ### 📈 Three Monitoring Levels
-- **📍 Detailed History**: Minute-by-minute recording of all connectivity tests
-- **⏰ 24-Hour Overview**: Hourly aggregation with availability percentages  
-- **📅 30-Day Overview**: Daily long-term trends for SLA monitoring
+- **📍 Detailed History**: Minute-by-minute recording of all connectivity tests including ping latency
+- **⏰ 24-Hour Overview**: Hourly aggregation with availability percentages plus average and 99th percentile ping
+- **📅 30-Day Overview**: Daily long-term trends including average and 99th percentile ping for SLA monitoring
 
 ### 🎯 Professional Availability Levels
 - 🟢 **Excellent** (≥99.9%): Enterprise-grade availability
@@ -35,9 +35,11 @@ open http://localhost:8000
 ### ⚡ Technical Highlights
 - **Real-time Updates**: Automatic chart refresh every 60 seconds
 - **Responsive Design**: Optimized for desktop, tablet, and mobile
+- **Toggleable Charts**: Switch between availability and ping views
 - **Persistent Storage**: SQLite database with automatic cleanup
 - **Multi-Platform**: Supports AMD64 and ARM64 (Raspberry Pi, Apple Silicon)
 - **High Performance**: Minimal resource usage through Alpine Linux
+- **Ping Latency Measurement**: ICMP ping to capture latency and display in dashboard
 
 ## ⚙️ Configuration
 

README.md

@@ -1,6 +1,6 @@
 # Internet-Verbindungsmonitor 🌐
 
-Ein containerisiertes Internet-Monitoring-Tool, das die Verfügbarkeit Ihrer Internetverbindung kontinuierlich überwacht und in einer übersichtlichen Web-Oberfläche mit professionellen Diagrammen visualisiert.
+Ein containerisiertes Internet-Monitoring-Tool, das die Verfügbarkeit Ihrer Internetverbindung sowie die Ping-Latenz kontinuierlich überwacht und in einer übersichtlichen Web-Oberfläche mit professionellen Diagrammen visualisiert.
 
 ![Internet Monitor Dashboard](https://img.shields.io/badge/Status-Production%20Ready-green)
 ![Container](https://img.shields.io/badge/Container-Podman%2FDocker-blue)
@@ -38,9 +38,9 @@ cd internet-check
 
 ### 📈 Drei Monitoring-Ebenen
 
-- **📍 Detaillierter Verlauf**: Minutengenaue Aufzeichnung aller Connectivity-Tests
-- **⏰ 24-Stunden-Übersicht**: Stündliche Aggregation mit Verfügbarkeitsprozenten
-- **📅 30-Tage-Übersicht**: Tägliche Langzeittrends für SLA-Monitoring
+- **📍 Detaillierter Verlauf**: Minutengenaue Aufzeichnung aller Connectivity-Tests inklusive Ping-Latenz
+- **⏰ 24-Stunden-Übersicht**: Stündliche Aggregation mit Verfügbarkeitsprozenten sowie durchschnittlicher und 99%-Perzentil-Ping
+- **📅 30-Tage-Übersicht**: Tägliche Langzeittrends inklusive durchschnittlicher und 99%-Perzentil-Ping-Latenz für SLA-Monitoring
 
 ### 🎯 Professionelle Verfügbarkeitsstufen
 
@@ -53,9 +53,11 @@ cd internet-check
 
 - **Echtzeit-Updates**: Automatische Diagramm-Aktualisierung alle 60 Sekunden
 - **Responsive Design**: Optimiert für Desktop, Tablet und Mobile
+- **Umschaltbare Diagramme**: Verfügbarkeit und Ping lassen sich getrennt anzeigen
 - **Persistente Speicherung**: SQLite-Datenbank mit automatischem Cleanup
 - **Container-Ready**: Podman/Docker-basiert ohne komplexe Abhängigkeiten
 - **Hochperformant**: Minimaler Ressourcenverbrauch durch Alpine Linux
+- **Ping-Latenzmessung**: ICMP-Ping zur Erfassung der aktuellen Latenz und Anzeige im Dashboard
 
 ## ⚙️ Konfiguration
 
@@ -161,25 +163,29 @@ Das Projekt ist jetzt modular strukturiert:
 
 ### 📊 Datenmodell
 
+Die Status-Tabelle speichert neben dem Onlinestatus auch die gemessene Ping-Latenz:
+
 ```sql
 CREATE TABLE status (
-    ts INTEGER PRIMARY KEY,  -- Unix-Timestamp (Sekunden seit 1970)
-    up INTEGER              -- 1=online, 0=offline
+    ts   INTEGER PRIMARY KEY,  -- Unix-Timestamp (Sekunden seit 1970)
+    up   INTEGER,              -- 1 = online, 0 = offline
+    ping REAL                  -- gemessene Ping-Latenz in Millisekunden
 );
 ```
 
 ### 🔗 API-Endpunkte
 
 - `GET /` - Responsive Web-Dashboard
-- `GET /data` - JSON-API mit Roh- und Aggregationsdaten
+- `GET /data` - JSON-API mit Roh- und Aggregationsdaten inklusive Ping-Latenzen (`pings`)
 
-### � Monitoring-Algorithmus
+### 🛰️ Monitoring-Algorithmus
 
-1. **HTTP-Test**: GET-Request an konfigurierte URL
-2. **Timeout-Handling**: 5 Sekunden maximale Wartezeit
-3. **Bewertung**: HTTP 2xx = Online | Timeout/Error = Offline
-4. **Speicherung**: Timestamp + Status in SQLite
-5. **Retention**: Automatisches Cleanup alter Daten
+1. **Ping-Messung**: ICMP-Ping zur Ermittlung der Latenz
+2. **HTTP-Test**: GET-Request an die konfigurierte URL
+3. **Timeout-Handling**: 5 Sekunden maximale Wartezeit
+4. **Bewertung**: HTTP 2xx = Online, sonst Offline
+5. **Speicherung**: Timestamp + Status + Ping in SQLite
+6. **Retention**: Automatisches Cleanup alter Daten
 
 ## 💡 Anwendungsfälle
 
@@ -214,19 +220,21 @@ podman exec monitor-primary cp /app/data.db /tmp/
 podman cp monitor-primary:/tmp/data.db ./backup_$(date +%Y%m%d).db
 
 # CSV-Export für Excel-Analyse
-podman exec monitor-primary sqlite3 /app/data.db \
-  "SELECT datetime(ts, 'unixepoch') as timestamp, 
-          CASE up WHEN 1 THEN 'Online' ELSE 'Offline' END as status 
-   FROM status ORDER BY ts;" \
-  -header -csv > connectivity_report.csv
+  podman exec monitor-primary sqlite3 /app/data.db \
+    "SELECT datetime(ts, 'unixepoch') as timestamp,
+            CASE up WHEN 1 THEN 'Online' ELSE 'Offline' END as status,
+            ping
+     FROM status ORDER BY ts;" \
+    -header -csv > connectivity_report.csv
 
 # Statistiken abrufen
-podman exec monitor-primary sqlite3 /app/data.db \
-  "SELECT 
-     COUNT(*) as total_checks,
-     SUM(up) as successful_checks,
-     ROUND(SUM(up) * 100.0 / COUNT(*), 2) as uptime_percentage
-   FROM status WHERE ts >= strftime('%s', 'now', '-24 hours');"
+  podman exec monitor-primary sqlite3 /app/data.db \
+    "SELECT
+       COUNT(*) as total_checks,
+       SUM(up) as successful_checks,
+       ROUND(SUM(up) * 100.0 / COUNT(*), 2) as uptime_percentage,
+       ROUND(AVG(ping), 2) as avg_ping_ms
+     FROM status WHERE ts >= strftime('%s', 'now', '-24 hours');"
 ```
 
 ## 📊 SLA-Referenztabelle

monitor.py

@@ -1,4 +1,5 @@
 import os, time, threading, sqlite3, datetime as dt, requests
+import subprocess, urllib.parse, re, math
 from flask import Flask, jsonify, Response, render_template
 
 # ────────── Konfiguration via ENV ──────────
@@ -11,27 +12,45 @@ def get_utc_timestamp():
     """UTC-Timestamp zurückgeben (immer UTC, unabhängig von Container-Zeitzone)"""
     return int(dt.datetime.now(dt.timezone.utc).timestamp())
 
+def ping_host():
+    """Einmaliges Ping auf das TEST_URL-Host ausführen und Laufzeit in ms zurückgeben."""
+    host = urllib.parse.urlparse(TEST_URL).hostname or TEST_URL
+    try:
+        r = subprocess.run(["ping", "-c", "1", "-W", "1", host], capture_output=True, text=True)
+        if r.returncode == 0:
+            m = re.search(r"time=([0-9.]+) ms", r.stdout)
+            if m:
+                return float(m.group(1))
+    except Exception:
+        pass
+    return None
+
 # ────────── Backend ──────────
 
 app = Flask(__name__)
 
 def init_db():
     """Tabelle anlegen, falls sie noch nicht existiert."""
     with sqlite3.connect(DB_PATH) as c:
-        c.execute("CREATE TABLE IF NOT EXISTS status(ts INTEGER PRIMARY KEY, up INTEGER)")
+        c.execute("CREATE TABLE IF NOT EXISTS status(ts INTEGER PRIMARY KEY, up INTEGER, ping REAL)")
+        try:
+            c.execute("ALTER TABLE status ADD COLUMN ping REAL")
+        except sqlite3.OperationalError:
+            pass
 
 def write_sample():
     """Einen Connectivity-Datensatz schreiben und alte Daten trimmen."""
+    ping_ms = ping_host()  # Ping immer messen, unabhängig vom HTTP-Test
     up = 1
-    try: 
+    try:
         requests.get(TEST_URL, timeout=5)
-    except requests.RequestException: 
+    except requests.RequestException:
         up = 0
 
     now = get_utc_timestamp()  # UTC-Timestamp verwenden
     with sqlite3.connect(DB_PATH) as c:
         # INSERT OR REPLACE verwenden um Duplikate zu vermeiden
-        c.execute("INSERT OR REPLACE INTO status VALUES(?,?)", (now, up))
+        c.execute("INSERT OR REPLACE INTO status(ts,up,ping) VALUES(?,?,?)", (now, up, ping_ms))
         # Alte Daten löschen - aber nur sehr alte (älter als RETENTION)
         cutoff = now - RETENTION*86400
         deleted = c.execute("DELETE FROM status WHERE ts < ?", (cutoff,)).rowcount
@@ -52,7 +71,7 @@ def index():
 def data():
     # Alle Daten holen (ohne since Filter) um das Problem zu debuggen
     with sqlite3.connect(DB_PATH) as c:
-        rows = c.execute("SELECT ts,up FROM status ORDER BY ts").fetchall()
+        rows = c.execute("SELECT ts,up,ping FROM status ORDER BY ts").fetchall()
         total_count = c.execute("SELECT COUNT(*) FROM status").fetchone()[0]
 
     print(f"DEBUG: Total records: {total_count}, All records: {len(rows)}")
@@ -73,30 +92,40 @@ def data():
     # Basis-Daten für detaillierten Verlauf - nur letzten 12 Stunden
     twelve_hours_ago = now - (12 * 3600)  # 12 Stunden in Sekunden
     detail_rows = [row for row in rows if row[0] >= twelve_hours_ago]
-    
+
     result = {
-        "labels":[t for t,_ in detail_rows],  # Raw UTC timestamps der letzten 12h senden
-        "values":[u for _,u in detail_rows]
+        "labels":[t for t,_,_ in detail_rows],  # Raw UTC timestamps der letzten 12h senden
+        "values":[u for _,u,_ in detail_rows],
+        "pings":[p for _,_,p in detail_rows]
     }
 
     # 24-Stunden-Aggregation (letzte 24h)
     current_hour_start = (now // 3600) * 3600  # Aktuelle Stunde, auf den Stundenanfang gerundet
     hourly_data = []
-    
+
     for h in range(24):
         # Von der aktuellen Stunde 23 Stunden zurückgehen
         hour_start = current_hour_start - (h * 3600)
         hour_end = hour_start + 3600
-        hour_rows = [up for ts, up in rows if hour_start <= ts < hour_end]
-        
+        hour_rows = [(up, ping) for ts, up, ping in rows if hour_start <= ts < hour_end]
+
         if hour_rows:
-            uptime = sum(hour_rows) / len(hour_rows)
+            ups = [u for u, _ in hour_rows]
+            pings = [p for _, p in hour_rows if p is not None]
+            uptime = sum(ups) / len(ups)
+            ping_avg = sum(pings) / len(pings) if pings else None
+            if pings:
+                k = max(0, min(len(pings)-1, math.ceil(len(pings) * 0.99) - 1))
+                ping_p99 = sorted(pings)[k]
+            else:
+                ping_p99 = None
         else:
             uptime = -1  # Keine Daten verfügbar
-
+            ping_avg = ping_p99 = None
+
         # Nur die Stunde des Tages (0-23) senden
         hour_of_day = dt.datetime.fromtimestamp(hour_start, dt.timezone.utc).hour
-        hourly_data.insert(0, {"hour": hour_of_day, "uptime": uptime})
+        hourly_data.insert(0, {"hour": hour_of_day, "uptime": uptime, "ping_avg": ping_avg, "ping_p99": ping_p99})
 
     # Tages-Aggregation (letzte 30 Tage) - korrekte Kalendertage verwenden
     daily_data = []
@@ -105,23 +134,32 @@ def data():
     for d in range(30):
         # Kalendertag berechnen (d Tage zurück vom heutigen Tag)
         target_date = current_utc_date - dt.timedelta(days=d)
-        
+
         # Start und Ende des Kalendertages in UTC
         day_start_dt = dt.datetime.combine(target_date, dt.time.min).replace(tzinfo=dt.timezone.utc)
         day_end_dt = dt.datetime.combine(target_date, dt.time.max).replace(tzinfo=dt.timezone.utc)
-        
+
         day_start = int(day_start_dt.timestamp())
         day_end = int(day_end_dt.timestamp())
-        
-        day_rows = [up for ts, up in rows if day_start <= ts <= day_end]
-        
+
+        day_rows = [(up, ping) for ts, up, ping in rows if day_start <= ts <= day_end]
+
         if day_rows:
-            uptime = sum(day_rows) / len(day_rows)
+            ups = [u for u, _ in day_rows]
+            pings = [p for _, p in day_rows if p is not None]
+            uptime = sum(ups) / len(ups)
+            ping_avg = sum(pings) / len(pings) if pings else None
+            if pings:
+                k = max(0, min(len(pings)-1, math.ceil(len(pings) * 0.99) - 1))
+                ping_p99 = sorted(pings)[k]
+            else:
+                ping_p99 = None
         else:
             uptime = -1  # Keine Daten verfügbar
-
+            ping_avg = ping_p99 = None
+
         # UTC-Timestamp für Tagesbeginn senden
-        daily_data.insert(0, {"date": day_start, "uptime": uptime})
+        daily_data.insert(0, {"date": day_start, "uptime": uptime, "ping_avg": ping_avg, "ping_p99": ping_p99})
 
     result["hourly"] = hourly_data
     result["daily"] = daily_data