Skip to main content
Account

Telegraf

Telegraf is an open-source agent for collecting, processing and forwarding metrics and logs. Learn how input, processor, aggregator and output plugins work.

Status
Stable
Category
Observability and Analysis
Subcategory
Observability
License
MIT
Technology matrix Skip Just not for me See the matrix

Telegraf is an open-source, plugin-driven agent from InfluxData for collecting, processing, and forwarding metrics, events, and logs. It ships as a single Go binary with no runtime dependencies and is commonly used as the scraping and shipping layer in front of a time-series database.

Its architecture is organized around four kinds of plugins declared in a TOML config. Input plugins gather data: system stats, SNMP, Kafka, Kubernetes, MySQL, PostgreSQL, Prometheus endpoints, cloud provider APIs, and hundreds more. Processor plugins enrich, filter, rename, or tag points in flight. Aggregator plugins compute rolling statistics like min/max/percentiles. Output plugins write the results to destinations such as InfluxDB, Prometheus remote write, Kafka, Elasticsearch, CloudWatch, Datadog, or files. Telegraf speaks InfluxDB Line Protocol natively but supports JSON, CSV, Prometheus, OpenTelemetry, and many other formats.

Because a single agent can ingest from many sources and fan out to many sinks, Telegraf is frequently used as a vendor-neutral metrics collector in hybrid environments, a protocol translator in front of Prometheus or OTLP pipelines, and the default on-host agent for InfluxDB and Grafana Cloud deployments.

Telegraf recordings

5 recordings, newest first

Observability

Email signup loads here. You can also manage email preferences in account settings.

Build Telegraf? I'll show you where engineers get stuck with it → Back free content for the community →