DocsOpenTelemetry
Send OpenTelemetry data
Televane accepts traces, metrics and logs in the OpenTelemetry protocol (OTLP). Any OpenTelemetry SDK or the OpenTelemetry Collector can send to it — there is no proprietary agent in your code.
1. Create an ingest token
- In the portal open Administration › API keys and create a token with the ingest scope.
- Copy it (it starts with
obs_). It is shown once — keep it in your secret store, not in source code.
2. Point your application at Televane
| Protocol | Endpoint |
|---|---|
| OTLP/HTTP (protobuf or JSON, optional gzip) | https://ingest.televane.com — paths /v1/traces, /v1/metrics, /v1/logs |
| OTLP/gRPC | ingest.televane.com:443 (TLS) |
| Authentication | Header Authorization: Bearer obs_… |
Most SDKs read the standard environment variables, so you rarely need to change code:
export OTEL_SERVICE_NAME=checkout-api
export OTEL_EXPORTER_OTLP_ENDPOINT=https://ingest.televane.com
export OTEL_EXPORTER_OTLP_PROTOCOL=http/protobuf
export OTEL_EXPORTER_OTLP_HEADERS="Authorization=Bearer obs_YOUR_TOKEN"
On Windows PowerShell use $env:OTEL_SERVICE_NAME="checkout-api" and so on. Set OTEL_SERVICE_NAME for every service: it is how services are named on the APM page and the service map.
3. Or use the OpenTelemetry Collector
If you already run a Collector, add Televane as an exporter:
exporters:
otlphttp/televane:
endpoint: https://ingest.televane.com
headers:
Authorization: "Bearer ${env:TELEVANE_TOKEN}"
service:
pipelines:
traces: { receivers: [otlp], processors: [batch], exporters: [otlphttp/televane] }
metrics: { receivers: [otlp], processors: [batch], exporters: [otlphttp/televane] }
logs: { receivers: [otlp], processors: [batch], exporters: [otlphttp/televane] }
What you will see
- APM — throughput, error rate and p95 latency per service.
- Traces — recent traces with their full span waterfall and the logs of the same request.
- Service map — services, databases, queues and external calls, with calls per minute, errors and latency on every edge.
- Custom metrics — your application metrics (gauges, counters, histograms and summaries), with labels, filters and group-by.
- Logs — searchable, linked to their traces.
Good to know
- Requests are limited to 1 MiB compressed (8 MiB uncompressed) and 5,000 spans or log records each; the SDKs' batch processors stay well within this.
- Passwords, tokens, keys, card numbers and SQL literals in attributes and log bodies are masked before storage.
- Exponential histograms are not supported yet; use explicit-bucket histograms.
- Data above your plan's monthly volume is rejected on Free (HTTP 429) and billed per GB on paid plans.