Self-hosted event collection.
Single binary. Zero dependencies.

Your data, your infrastructure. One Go binary collects events from Snowplow trackers, webhooks, and tracking pixels — then delivers them to your warehouse.

docker pull ghcr.io/deepskydatahq/straumheim:latest
Get Started View on GitHub

Everything you need, nothing you don't

Single Binary

One Go binary, one YAML config. Deploy with docker run. No Kafka, no Redis, no queues.

Multi-Protocol

Snowplow trackers, webhooks, and tracking pixels. One collector for all your event sources.

Zero Dependencies

No infrastructure requirements beyond the binary itself. Runs on a $5/month VPS with unlimited events.

Warehouse Delivery

Batch delivery to Postgres (COPY), ClickHouse (HTTP), JSONL files. Auto-schema evolution.

From zero to collecting events in minutes

1

Install

terminal
$ docker pull ghcr.io/deepskydatahq/straumheim:latest
2

Configure

config.yaml
server:
  port: 8080

inputs:
  webhook:
    enabled: true
    path: /webhook

sinks:
  postgres:
    enabled: true
    connection_string: postgres://localhost:5432/events
    table: events
    batch_size: 1000
3

Collect events

terminal
$ curl -X POST http://localhost:8080/webhook \
    -H "Content-Type: application/json" \
    -d '{"event": "signup", "user_id": "abc123"}'

{"status":"ok","events_received":1}
4

Query your data

psql
SELECT id, protocol, payload FROM events;

       id       | protocol |              payload
----------------+----------+-----------------------------------
 ev_1a2b3c4d   | webhook  | {"event":"signup","user_id":"abc123"}
(1 row)

How Straumheim compares

Segment Snowplow OSS Straumheim
Cost $120/mo for 10K MTU Free (self-hosted) Free (self-hosted)
Infrastructure Managed (vendor) Kafka/Kinesis + Spark/Beam Single binary
Setup time Minutes Days/weeks Minutes
Data ownership Third-party servers Your infrastructure Your infrastructure
Tracker support Segment SDKs only Snowplow SDKs Snowplow SDKs + webhooks + pixels
Destinations Limited free, paid add-ons S3/GCS/Redshift Postgres, ClickHouse, JSONL, stdout