_
_
_
_
Glances Cheatsheet¶
• Installation
| Platform | Command |
|---|---|
| Ubuntu/Debian | INLINE_CODE_10 |
| RHEL/CentOS/Fedora | INLINE_CODE_11 or INLINE_CODE_12 |
| Arch Linux | INLINE_CODE_13 |
| macOS (Homebrew) | INLINE_CODE_14 |
| Windows (pip) | INLINE_CODE_15 |
| Windows (Chocolatey) | INLINE_CODE_16 |
| Docker | INLINE_CODE_17 |
| Python pip (all platforms) | INLINE_CODE_18 |
| With all features | INLINE_CODE_19 |
| From source | INLINE_CODE_20 |
oder Grundlegende Befehle
| Command | Description |
|---|---|
| INLINE_CODE_21 | Launch glances with default settings |
| INLINE_CODE_22 | Set refresh interval to 5 seconds |
| INLINE_CODE_23 | Display version information |
| INLINE_CODE_24 | Show help and available options |
| INLINE_CODE_25 | Show per-CPU statistics by default |
| INLINE_CODE_26 | Show only active processes with short names |
| INLINE_CODE_27 | Filter processes by name (e.g., "python") |
| INLINE_CODE_28 | Show processes for specific user |
| INLINE_CODE_29 | Use white theme (black background) |
| INLINE_CODE_30 | Disable background colors |
| INLINE_CODE_31 | Display temperature in Fahrenheit instead of Celsius |
| INLINE_CODE_32 | Display network rate in bytes per second |
| INLINE_CODE_33 | Limit displayed processes to 50 |
| INLINE_CODE_34 | Hide kernel threads from process list |
| INLINE_CODE_35 | Disable bold text in interface |
| _ | |
| Interaktive Tastenkombinationen |
| Key | Description |
|---|---|
| INLINE_CODE_36 or INLINE_CODE_37 | Quit glances |
| INLINE_CODE_38 | Display help screen |
| INLINE_CODE_39 | Sort processes by CPU usage |
| INLINE_CODE_40 | Sort processes by memory usage |
| INLINE_CODE_41 | Sort processes by process name |
| INLINE_CODE_42 | Sort processes by I/O rate |
| INLINE_CODE_43 | Sort processes by process time |
| INLINE_CODE_44 | Sort processes by user |
| INLINE_CODE_45 | Kill selected process (requires confirmation) |
| INLINE_CODE_46 | Renice (change priority) selected process |
| INLINE_CODE_47 | Toggle between global CPU and per-CPU statistics |
| INLINE_CODE_48 | Show/hide disk I/O stats |
| INLINE_CODE_49 | Show/hide filesystem stats |
| INLINE_CODE_50 | Show/hide network stats |
| INLINE_CODE_51 | Show/hide sensors |
| INLINE_CODE_52 | Show/hide left sidebar |
| INLINE_CODE_53 | Show/hide processes list |
| INLINE_CODE_54 | Toggle network rate between bits/s and bytes/s |
| INLINE_CODE_55 | Delete warning logs |
| INLINE_CODE_56 | Delete warning and critical logs |
| INLINE_CODE_57 | Show/hide Docker containers |
| INLINE_CODE_58 | Toggle process tree view |
| INLINE_CODE_59 | Navigate through process list |
| INLINE_CODE_60 | Tag/untag process for bulk operations |
| _ | |
| / Fortgeschrittene Nutzung |
| Command | Description |
|---|---|
| INLINE_CODE_61 | Start in server mode (default port 61209) |
| INLINE_CODE_62 | Start server on custom port 8080 |
| INLINE_CODE_63 | Start server with password authentication |
| INLINE_CODE_64 | Bind server to specific IP address |
| INLINE_CODE_65 | Connect to remote glances server |
| INLINE_CODE_66 | Connect to server on custom port |
| INLINE_CODE_67 | Start web server mode (default port 61208) |
| INLINE_CODE_68 | Start web server on custom port |
| INLINE_CODE_69 | Web server with authentication |
| INLINE_CODE_70 | Make web server accessible from all interfaces |
| INLINE_CODE_71 | Enable Docker container monitoring |
| INLINE_CODE_72 | Export metrics to CSV file |
| INLINE_CODE_73 | Export metrics to InfluxDB |
| INLINE_CODE_74 | Export metrics for Prometheus scraping |
| INLINE_CODE_75 | Export metrics to JSON file |
| INLINE_CODE_76 | Disable specific plugins |
| INLINE_CODE_77 | Start SNMP server mode |
| INLINE_CODE_78 | Connect via SNMP |
| INLINE_CODE_79 | Log output to file |
| INLINE_CODE_80 | Enable debug logging |
| _ | |
| / Docker Nutzung |
Run Glances in Docker Container¶
# Basic container with web interface
docker run -d --restart="always" \
-p 61208-61209:61208-61209 \
-e GLANCES_OPT="-w" \
-v /var/run/docker.sock:/var/run/docker.sock:ro \
--pid host \
nicolargo/glances
# With custom configuration file
docker run -d \
-v /path/to/glances.conf:/glances/conf/glances.conf \
-p 61208-61209:61208-61209 \
-e GLANCES_OPT="-w" \
nicolargo/glances
# Server mode only
docker run -d \
-p 61209:61209 \
-e GLANCES_OPT="-s" \
--pid host \
nicolargo/glances
Konfiguration
Glances verwendet eine Konfigurationsdatei unter:
- Linux: ~/.config/glances/glances.conf oder /etc/glances/glances.conf_
- macOS: ~/Library/Application Support/glances/glances.conf
- Windows: %APPDATA%\glances\glances.conf_
Basic Configuration Beispiel¶
[global]
# Refresh rate in seconds
refresh=2
# Check for updates
check_update=false
[cpu]
# CPU thresholds (%)
careful=50
warning=70
critical=90
[mem]
# Memory thresholds (%)
careful=50
warning=70
critical=90
[memswap]
# Swap thresholds (%)
careful=50
warning=70
critical=90
[load]
# Load average thresholds (per CPU)
careful=0.7
warning=1.0
critical=5.0
[network]
# Hide network interfaces
hide=lo,docker.*
# Show cumulative values
cumulative=False
[diskio]
# Hide disk devices
hide=loop.*,/dev/loop.*
[fs]
# Hide filesystems
hide=/boot.*,/snap.*
# Careful threshold (%)
careful=50
warning=70
critical=90
[folders]
# Monitor specific folders
folder_1_path=/home
folder_1_careful=2500
folder_1_warning=3000
folder_1_critical=3500
[sensors]
# Temperature thresholds (Celsius)
temperature_core_careful=60
temperature_core_warning=70
temperature_core_critical=80
[processlist]
# CPU thresholds for processes (%)
cpu_careful=50
cpu_warning=70
cpu_critical=90
# Memory thresholds for processes (%)
mem_careful=50
mem_warning=70
mem_critical=90
[docker]
# Enable Docker monitoring
disable=False
# Maximum containers to display
max_name_size=20
Web Server Konfiguration¶
[outputs]
# Export to CSV
export_csv_file=/tmp/glances.csv
export_csv_overwrite=False
# Export to InfluxDB
export_influxdb=True
influxdb_host=localhost
influxdb_port=8086
influxdb_user=root
influxdb_password=root
influxdb_db=glances
# Export to Prometheus
export_prometheus=True
prometheus_port=9091
prometheus_prefix=glances
[serverlist]
# Remote servers to monitor
server_1_name=prod-server-1
server_1_alias=Production 1
server_1_host=192.168.1.100
server_1_port=61209
server_2_name=prod-server-2
server_2_alias=Production 2
server_2_host=192.168.1.101
server_2_port=61209
 RESTful API Nutzung
API Endpoints¶
| Endpoint | Description |
|---|---|
| INLINE_CODE_85 | Get server status |
| INLINE_CODE_86 | Get server configuration |
| INLINE_CODE_87 | Get all stats in JSON format |
| INLINE_CODE_88 | Get CPU statistics |
| INLINE_CODE_89 | Get memory statistics |
| INLINE_CODE_90 | Get swap memory statistics |
| INLINE_CODE_91 | Get load average |
| INLINE_CODE_92 | Get network statistics |
| INLINE_CODE_93 | Get disk I/O statistics |
| INLINE_CODE_94 | Get filesystem statistics |
| INLINE_CODE_95 | Get process list |
| INLINE_CODE_96 | Get sensors data |
| INLINE_CODE_97 | Get Docker containers stats |
| INLINE_CODE_98 | Get specific process info |
| _ | |
| ### API Beispiele |
# Get CPU stats
curl http://localhost:61208/api/3/cpu
# Get memory stats
curl http://localhost:61208/api/3/mem
# Get all stats
curl http://localhost:61208/api/3/all
# Get specific process
curl http://localhost:61208/api/3/processlist/pid/1234
# Pretty print JSON output
curl http://localhost:61208/api/3/cpu | python -m json.tool
# Get stats with authentication
curl -u admin:secret http://localhost:61208/api/3/cpu
Häufige Anwendungsfälle
Use Case 1: Remote Server Monitoring¶
Überwachen Sie mehrere Server von einem zentralen Standort:
# On remote servers (run as service)
glances -s -B 0.0.0.0 --password
# On monitoring station
glances -c server1.example.com --password yourpassword
# Or use web interface
glances -w -B 0.0.0.0 --username admin --password secret
# Access via browser: http://server:61208
Use Case 2: Export von Metriken nach InfluxDB und Grafana¶
Kontinuierliche Überwachung mit Visualisierung:
# Start glances with InfluxDB export
glances --export influxdb \
--influxdb-host localhost \
--influxdb-port 8086 \
--influxdb-user admin \
--influxdb-password secret \
--influxdb-db glances
# Or configure in glances.conf and run as daemon
glances --quiet --export influxdb
Konfigurieren Sie Grafana, um InfluxDB als Datenquelle zu verwenden und Dashboards zu erstellen.
Use Case 3: Docker Container Monitoring¶
Monitor Docker Container neben Systemmetriken:
# Enable Docker monitoring
glances --docker
# Docker-only view (hide processes)
glances --docker --disable-process
# Export Docker stats to Prometheus
glances --docker --export prometheus --prometheus-port 9091
# Web interface with Docker stats
glances -w --docker
Use Case 4: Fehlerbehebung Hohe CPU Verwendung¶
Identifizieren und analysieren ressourcenintensive Prozesse:
# Start glances with focus on CPU
glances --percpu
# Interactive steps:
# 1. Press 'c' to sort by CPU usage
# 2. Press '1' to see per-CPU breakdown
# 3. Press 'T' to view process tree
# 4. Navigate to problematic process
# 5. Press 'k' to kill if needed
# Export data for analysis
glances --export csv --export-csv-file /tmp/cpu-analysis.csv -t 1
Use Case 5: Automatische Überwachung Script¶
Erstellen Sie ein Überwachungsskript mit Warnungen:
#!/bin/bash
# monitor.sh - Check system health via Glances API
# Start glances web server in background
glances -w --quiet &
GLANCES_PID=$!
sleep 5
# Check CPU usage
CPU_USAGE=$(curl -s http://localhost:61208/api/3/cpu | \
python -c "import sys, json; print(json.load(sys.stdin)['total'])")
if (( $(echo "$CPU_USAGE > 80" | bc -l) )); then
echo "ALERT: CPU usage is ${CPU_USAGE}%"
# Send notification or alert
fi
# Check memory usage
MEM_USAGE=$(curl -s http://localhost:61208/api/3/mem | \
python -c "import sys, json; print(json.load(sys.stdin)['percent'])")
if (( $(echo "$MEM_USAGE > 90" | bc -l) )); then
echo "ALERT: Memory usage is ${MEM_USAGE}%"
fi
# Cleanup
kill $GLANCES_PID
oder Best Practices
-
** Verwenden Sie Konfigurationsdateien** anstelle von Befehlszeilenoptionen für permanente Einstellungen, um Konsistenz über Sitzungen zu erhalten und Konfigurationen versionskontrollierbar zu machen
-
** Setzen Sie geeignete Refresh-Intervalle** - Verwenden Sie 2-5 Sekunden zur interaktiven Überwachung, 10-30 Sekunden zur Hintergrundüberwachung, um die Reaktionsfähigkeit mit Systemlast auszugleichen
-
Filter unnötige Informationen - Verstecken irrelevanter Netzwerkschnittstellen, Festplattengeräte und Dateisysteme, um Cutter zu reduzieren und sich auf wichtige Metriken zu konzentrieren
-
Implementieren Sie die Authentifizierung für Fernüberwachung und Webschnittstellen, um unberechtigten Zugriff auf Systeminformationen zu verhindern, insbesondere in Produktionsumgebungen
-
**Export zu Zeitreihen-Datenbanken* wie InfluxDB oder Prometheus für langfristige Trendanalyse, Kapazitätsplanung und historische Fehlerbehebung
-
Run als systemd service auf Linux-Servern für automatische Inbetriebnahme und Überwachung Kontinuität über Neustarts
-
Monitor Docker-Container beim Betrieb von Container-Workloads, um die Ressourcennutzung pro Container zu verfolgen und ressourcenhungrige Anwendungen zu identifizieren
-
** Verwenden Sie Farbschwellen weise* - Konfigurieren Sie vorsichtige/warning/kritische Schwellen basierend auf den normalen Betriebsbereichen Ihres Systems, um Alarmermüdung zu vermeiden
-
Verwalten Sie die API für Automatisierung und Integration mit vorhandenen Überwachungssystemen, Warnwerkzeugen oder benutzerdefinierten Dashboards
-
**Keep Blicke aktualisiert* - Regelmäßig aktualisieren, um neue Features, Leistungsverbesserungen und Sicherheits-Patches zu erhalten
Fehlerbehebung
| Issue | Solution |
|---|---|
| INLINE_CODE_99 | Ensure glances is installed and in PATH. Try INLINE_CODE_100 and add INLINE_CODE_101 to PATH |
| High CPU usage by glances itself | Increase refresh interval with INLINE_CODE_102 or higher. Disable unused plugins with INLINE_CODE_103 |
| Cannot connect to remote server | Check firewall rules allow port 61209. Verify server is running with INLINE_CODE_104. Test connectivity with INLINE_CODE_105 |
| Web interface not accessible | Ensure glances started with INLINE_CODE_106 flag. Check if port 61208 is available with INLINE_CODE_107. Bind to all interfaces with INLINE_CODE_108 |
| Docker containers not showing | Verify Docker socket is accessible: INLINE_CODE_109. Run glances with appropriate permissions or add user to docker group |
| Permission denied errors | Run with INLINE_CODE_110 for full system access, or adjust permissions for specific resources. Some features require root privileges |
| Sensors not displaying | Install INLINE_CODE_111 package and run INLINE_CODE_112. Install Python sensors library: INLINE_CODE_113 |
| InfluxDB export not working | Verify InfluxDB is running and accessible. Check credentials and database name. Install InfluxDB client: INLINE_CODE_114 |
| Missing GPU information | Install NVIDIA monitoring library: INLINE_CODE_115. Ensure NVIDIA drivers are installed and working |
| Config file changes not applied | Verify config file location with INLINE_CODE_116. Check file syntax. Restart glances after configuration changes |
| Process list incomplete | Increase process limit with INLINE_CODE_117 for unlimited. Check if kernel threads are hidden with INLINE_CODE_118 |
| Unicode/display issues in terminal | Set proper locale: INLINE_CODE_119. Use INLINE_CODE_120 flag if terminal doesn't support Unicode |
| _ | |
| Optionale Abhängigkeiten |
Installieren Sie diese Pakete für verbesserte Funktionalität:
# Docker monitoring
pip3 install docker
# GPU monitoring (NVIDIA)
pip3 install nvidia-ml-py3
# Sensors support (Linux)
pip3 install py3sensors
# Web interface
pip3 install bottle
# InfluxDB export
pip3 install influxdb influxdb-client
# Prometheus export
pip3 install prometheus_client
# Elasticsearch export
pip3 install elasticsearch
# StatsD export
pip3 install statsd
# IP geolocation
pip3 install geoip2
# All optional features
pip3 install 'glances[all]'