Infrastructure Monitoring Basics with Telegraf, Grafana and InfluxDB
summary
Step-by-step build of a hybrid monitoring stack that collects metrics with Telegraf, stores everything in InfluxDB 3.0 and visualizes with Grafana, using Prometheus scraping, SNMP for routers, Nvidia SMI for GPUs and CloudWatch for AWS; claims InfluxDB handles metrics logs and traces in one store at high ingest rates and supports SQL via Flight SQL; Telegraf contributes 300+ plugins and writes line protocol to InfluxDB; demo generates traces from Hot Rod, queries them via Jaeger-to-SQL and shows them in Grafana, and edge replication plus retention/downsampling features are presented as working or imminent.
AI-written summary · about
chapters (22)
- 0:00Intro
- 2:11Agenda
- 3:20Monitoring vs Observability
- 6:13Monitoring + Observability Fields
- 9:44Break it down
- 11:07Data Collection
- 11:37Input Plugins
- 12:51Telegraf Architecture.
- 13:37Telegraf Setup
- 15:01Telegraf sidecar
- 15:54Telegraf Config
- 17:19Input plugin configs
- 18:46Output Plugins
- 19:13Data Storage
- 20:16Concepts: Data Model
- 24:13OpenTelemetry Schema
- 26:37Grafana Flow
- 28:11Useful Queries
- 29:14Example Dashboard
- 29:43Alerting
- 32:54Whisper GPT Solution
- 33:33Try it yourself - Quick Starts
related talks
35mReal-Time Monitoring with Grafana, StatsD and InfluxDB
42mMonitoring Mesos with InfluxDB & the TICK Stack
42mDistributed Monitoring: How to understand the Chaos
25mTelegraf – A data collection agent
34mThe Grafana Stack: Building a Highly Available, Scalable Observability Platform- 42mDistributed Monitoring: How to understand the Chaos
43mDistributed Monitoring: How to Understand the Chaos
06mFrom Sensors to Insights: Building an Open Observability Pipeline for Industrial Systems