Skip to content

telegrafcheatsheet.md

Latest commit

 

History

History
310 lines (235 loc) · 8.55 KB

telegrafcheatsheet.md

File metadata and controls

310 lines (235 loc) · 8.55 KB
Raw
Title 📈 Telegraf
Group Monitoring
Icon 📈
Order 6
tags
monitoring
sysadmin
linux

Telegraf Sysadmin Cheatsheet

Telegraf is an open-source, plugin-driven server agent from InfluxData for collecting, processing, aggregating, and writing metrics. First released in 2015, it is part of the TICK stack (Telegraf, InfluxDB, Chronograf, Kapacitor) and supports 300+ input/output plugins.

Common use cases / Типичные сценарии: System metrics collection (CPU, memory, disk, network), application monitoring (MySQL, PostgreSQL, Redis, Nginx), SNMP polling, Docker/Kubernetes metrics, cloud metrics (AWS CloudWatch, Azure Monitor), custom script execution, Prometheus metrics scraping.

Status / Статус: Actively developed by InfluxData. Telegraf is one of the most versatile metrics collection agents. Alternatives include Prometheus node_exporter (Prometheus ecosystem), Grafana Alloy (Grafana LGTM stack), collectd (legacy, C-based), Vector (Rust-based, high-performance), vmagent (VictoriaMetrics ecosystem).

Default ports / Порты по умолчанию: None (outbound agent) | InfluxDB: 8086 | Prometheus metrics endpoint: 9273

📚 Table of Contents

1. Installation & Configuration

Install Telegraf

# RHEL/CentOS/AlmaLinux
cat > /etc/yum.repos.d/influxdb.repo << 'EOF'
[influxdb]
name=InfluxDB Repository
baseurl=https://repos.influxdata.com/rhel/$releasever/$basearch/stable
enabled=1
gpgcheck=1
gpgkey=https://repos.influxdata.com/influxdata-archive_compat.key
EOF
dnf install telegraf

# Debian/Ubuntu
wget -qO- https://repos.influxdata.com/influxdata-archive_compat.key | apt-key add -
echo "deb https://repos.influxdata.com/debian stable main" > /etc/apt/sources.list.d/influxdb.list
apt update && apt install telegraf

Main Configuration

/etc/telegraf/telegraf.conf

# Global agent settings
[agent]
  interval = "10s"                    # Collection interval / Интервал сбора
  round_interval = true               # Round collection to interval / Округлить до интервала
  metric_batch_size = 1000            # Batch size for output / Размер пакета
  metric_buffer_limit = 10000         # Buffer size / Размер буфера
  flush_interval = "10s"              # Flush interval / Интервал сброса
  flush_jitter = "0s"                 # Jitter for flush / Разброс сброса
  hostname = ""                       # Auto-detect / Авто-определение
  omit_hostname = false

# Output to InfluxDB v2
[[outputs.influxdb_v2]]
  urls = ["http://<INFLUXDB_HOST>:8086"]
  token = "<TOKEN>"
  organization = "<ORG>"
  bucket = "<BUCKET>"

# Output to Prometheus
# [[outputs.prometheus_client]]
#   listen = ":9273"
#   metric_version = 2

# System inputs (default)
[[inputs.cpu]]
  percpu = true
  totalcpu = true
  collect_cpu_time = false

[[inputs.disk]]
  ignore_fs = ["tmpfs", "devtmpfs", "devfs", "iso9660", "overlay", "aufs", "squashfs"]

[[inputs.diskio]]
[[inputs.mem]]
[[inputs.net]]
[[inputs.processes]]
[[inputs.swap]]
[[inputs.system]]

Output Plugin Comparison

Output Protocol Best For / Лучше для
influxdb_v2 HTTP/HTTPS InfluxDB v2 / TSDB
prometheus_client HTTP Prometheus scraping / Сбор Prometheus
elasticsearch HTTP/HTTPS ELK stack / Стек ELK
kafka TCP High-throughput pipeline / Высоконагруженный пайплайн
file Filesystem Debug/local storage / Отладка
http HTTP POST Custom webhooks / Кастомные вебхуки

2. Core Management

Generate Configuration

# Generate full default config
telegraf config > /etc/telegraf/telegraf.conf.default

# Generate config with specific plugins only
telegraf config --input-filter cpu:mem:disk:net --output-filter influxdb_v2 > /etc/telegraf/telegraf.conf

# List available input plugins
telegraf --input-list

# List available output plugins
telegraf --output-list

# Print sample config for specific plugin
telegraf --usage cpu
telegraf --usage influxdb_v2

Popular Input Plugins

# Nginx status
[[inputs.nginx]]
  urls = ["http://localhost/nginx_status"]

# MySQL statistics
[[inputs.mysql]]
  servers = ["<USER>:<PASSWORD>@tcp(<HOST>:3306)/"]

# PostgreSQL / PostgreSQL
[[inputs.postgresql]]
  address = "host=<HOST> user=<USER> password=<PASSWORD> sslmode=disable dbname=<DB>"

# Docker containers
[[inputs.docker]]
  endpoint = "unix:///var/run/docker.sock"

# HTTP response check
[[inputs.http_response]]
  urls = ["https://<HOST>"]
  response_timeout = "5s"

# SNMP polling
[[inputs.snmp]]
  agents = ["udp://<HOST>:161"]
  community = "<COMMUNITY_STRING>"
  [[inputs.snmp.field]]
    oid = ".1.3.6.1.2.1.1.3.0"
    name = "uptime"

# Execute custom script
[[inputs.exec]]
  commands = ["/usr/local/bin/custom_metric.sh"]
  data_format = "influx"
  timeout = "5s"
  interval = "60s"

3. Sysadmin Operations

Service Management

systemctl start telegraf      # Start / Запустить
systemctl stop telegraf       # Stop / Остановить
systemctl restart telegraf    # Restart / Перезапустить
systemctl enable telegraf     # Enable on boot / Автозапуск
systemctl status telegraf     # Check status / Проверить статус

Important Paths

Path Description / Описание
/etc/telegraf/telegraf.conf Main configuration / Основной конфиг
/etc/telegraf/telegraf.d/ Drop-in config fragments / Фрагменты конфигов
/var/log/telegraf/ Telegraf logs / Логи Telegraf

Tip

Use /etc/telegraf/telegraf.d/ to split configs into per-service files (e.g., nginx.conf, mysql.conf). Telegraf loads all .conf files from this directory. / Разделяйте конфиг на файлы по сервисам в /etc/telegraf/telegraf.d/.

Test Configuration

# Test config syntax
telegraf --config /etc/telegraf/telegraf.conf --test

# Dry run (collect once, print to stdout)
telegraf --config /etc/telegraf/telegraf.conf --once --test

# Run with debug output
telegraf --config /etc/telegraf/telegraf.conf --debug

4. Security

Environment Variables

/etc/default/telegraf (Debian) or /etc/sysconfig/telegraf (RHEL)

# Store secrets in environment
INFLUX_TOKEN=<TOKEN>
DB_PASSWORD=<PASSWORD>
# Reference in telegraf.conf
[[outputs.influxdb_v2]]
  token = "$INFLUX_TOKEN"

5. Troubleshooting & Tools

Common Issues

1. No Data in Output

# Test specific input
telegraf --config /etc/telegraf/telegraf.conf --input-filter cpu --test

# Test output connectivity
telegraf --config /etc/telegraf/telegraf.conf --output-filter influxdb_v2 --test

# Check logs
journalctl -u telegraf -f --no-pager

2. Plugin Permission Errors

# Add telegraf user to required groups
usermod -aG docker telegraf     # For Docker input / Для Docker
usermod -aG systemd-journal telegraf  # For journal input / Для journald

systemctl restart telegraf

3. High CPU/Memory Usage

# Reduce collection interval
# agent.interval = "30s" instead of "10s"

# Reduce buffer
# agent.metric_buffer_limit = 5000

# Check which plugins are slow
telegraf --config /etc/telegraf/telegraf.conf --debug 2>&1 | grep "took"

6. Logrotate Configuration

/etc/logrotate.d/telegraf

/var/log/telegraf/*.log {
    daily
    rotate 7
    compress
    delaycompress
    missingok
    notifempty
    create 640 telegraf telegraf
    postrotate
        systemctl kill --signal=HUP telegraf 2>/dev/null || true
    endscript
}

Documentation Links