StandardsInfluxDB line protocol endpoint

Telegraf Integration

Forward InfluxDB line protocol metrics from Telegraf agents to TigerOps. Monitor 200+ input plugins across your entire fleet with AI-powered anomaly detection and unified dashboards.

Connect Telegraf Book a Demo

Setup

How It Works

Install Telegraf Agent

Install Telegraf on your hosts using the official packages or Docker image. Enable the input plugins you need — CPU, memory, disk, network, and 200+ more are supported out of the box.

Add TigerOps Output Plugin

Add the [[outputs.http]] plugin to your telegraf.conf pointing to the TigerOps InfluxDB line protocol endpoint. Set your API key in the Authorization header.

Validate Metric Flow

Run telegraf --test to verify the output plugin can reach TigerOps. Check the live metric stream in your TigerOps workspace to confirm tags and field values are correct.

Configure Alerts and Dashboards

TigerOps auto-creates suggested dashboards based on detected Telegraf input plugins. Enable AI anomaly alerts per metric or across a fleet of Telegraf agents.

Capabilities

What You Get Out of the Box

InfluxDB Line Protocol Ingestion

Native HTTP endpoint accepting InfluxDB line protocol (ILP) v1 and v2. All Telegraf measurement names, tags, and fields are preserved and queryable without schema configuration.

200+ Input Plugin Support

Any Telegraf input plugin that emits line protocol works automatically — from system metrics (cpu, mem, disk) to databases (mysql, postgresql, redis) and cloud services.

Fleet-Wide Agent Management

Tag Telegraf agents by environment, region, and role. TigerOps aggregates metrics across your entire fleet for fleet-wide dashboards and per-host drill-down in a single view.

Aggregator & Processor Passthrough

Telegraf aggregator and processor plugins (basicstats, derivative, starlark) run before forwarding. TigerOps receives pre-aggregated data at any interval you configure.

Dynamic Tag-Based Alerting

Write alert rules that match Telegraf tag sets (host, region, service). A single rule covers your entire fleet, with per-host anomaly baselines computed independently.

Telegraf Config Versioning

Store and version your telegraf.conf files in TigerOps config management. Push configuration updates to agent fleets and roll back instantly if metric collection breaks.

Configuration

telegraf.conf Output Plugin

Add the TigerOps output section to your existing telegraf.conf file.

telegraf.conf

[agent]
  interval = "10s"
  flush_interval = "10s"
  metric_batch_size = 5000
  metric_buffer_limit = 50000
  hostname = ""          # auto-detected

# --- Inputs (examples) ---
[[inputs.cpu]]
  percpu = true
  totalcpu = true
  collect_cpu_time = false

[[inputs.mem]]
[[inputs.disk]]
  ignore_fs = ["tmpfs", "devtmpfs"]

[[inputs.diskio]]
[[inputs.net]]

# --- TigerOps Output ---
[[outputs.http]]
  url = "https://ingest.atatus.net/telegraf/write"
  data_format = "influx"
  timeout = "10s"
  [outputs.http.headers]
    Authorization = "Bearer ${TIGEROPS_API_KEY}"
    Content-Type  = "text/plain; charset=utf-8"

# Optional: tag all metrics with environment
[global_tags]
  env    = "production"
  region = "us-east-1"
  team   = "platform"

FAQ

Common Questions

Which Telegraf output plugin should I use with TigerOps?

Use [[outputs.http]] with data_format = "influx" pointing to https://ingest.atatus.net/telegraf/write. Alternatively, use [[outputs.influxdb_v2]] with the TigerOps endpoint and your API key as the token. Both work identically.

Can TigerOps handle high-cardinality Telegraf tag sets?

Yes. TigerOps is designed for high-cardinality environments. Tags like container_id or request_id are stored in the columnar store and indexed for fast filtering without the cardinality limits of traditional TSDBs.

Does TigerOps support Telegraf batch writes?

Yes. Set metric_batch_size and metric_buffer_limit in your Telegraf output plugin. TigerOps accepts batches up to 10 MB per request and processes them atomically to avoid partial ingestion.

Can I run multiple Telegraf agents on the same host?

Yes. Use distinct agent.hostname or add a custom global_tags.agent_id to disambiguate agents from the same host. TigerOps stores all agents independently and lets you filter by any tag combination.

How do I monitor Telegraf agent health itself?

Enable the [[inputs.internal]] plugin in Telegraf to expose agent metrics (gather_duration, write_duration, metrics_gathered). TigerOps will automatically surface agent health alongside your workload metrics.

Get Started

Turn Your Telegraf Fleet Into a Unified Observability Platform

One output plugin. 200+ input plugins. AI anomaly detection across your entire fleet. Deploy in 5 minutes.

Start Free Talk to an Engineer