Troubleshooting ingestion
It returns 200 but nothing shows, 401s, 413s and gRPC pitfalls — symptom, cause and fix.
opentelemetrytroubleshooting401413ingestion
First, split the problem
A 2xx from the endpoint means VerOps accepted and stored the data. If you then can't see it, the issue is almost always a missing resource attribute (so it isn't grouped into an entity) or you're looking at the wrong view / tenant / time range — not delivery. Confirm delivery with the smoke test in Quickstart, then use the table below.
Symptom → cause → fix
| Symptom | Likely cause | Fix |
|---|---|---|
Endpoint returns 200, data not on Servers | Metrics have no host.name | Add the resourcedetection processor (detectors [env, system]) to the metrics pipeline. |
200, not on Containers | No container-level metrics, or missing container.name | Run docker_stats (standalone) or kubeletstats (k8s) — both set container.name. |
Data labelled unknown-service | service.name not set | Set OTEL_SERVICE_NAME or a resource processor attribute. |
401 Unauthorized | Missing / wrong / placeholder key | Use a real key from Settings → API Keys in Authorization: Bearer …. Don't leave ta_live_YOUR_API_KEY. |
413 Payload Too Large (often on /v1/logs) | Batch bodies exceed the edge limit | Set send_batch_max_size: 8192 in the batch processor; enable gzip. |
Nothing arrives; exporter logs connection refused / no such host / deadline exceeded | Egress / DNS from the box, or gRPC to a public host | Confirm the host can reach https://ingest.verops.io; use http/protobuf, not gRPC, over the internet. |
| Collector runs but sends nothing | Only an otlp receiver and nothing pushing into it | Add a data source (hostmetrics / docker_stats / kubeletstats) or point apps at the Collector. |
| Config change had no effect | Collector not restarted | docker restart <collector> — a file edit alone doesn't reload it. |
Read the Collector logs
docker logs <collector> --tail=100Everything is ready. Begin running and processing data— started clean.error decoding …/unknown type— a config error; the Collector isn't running.Exporting failed …with401/413— key or batch-size problem (see table).
Confirm you're looking in the right place
- Tenant selector on the organization the key belongs to.
- Time range covering now (Last 15 minutes).
- Metrics live in Explore → Metrics / Data Explorer; hosts and containers in Infrastructure once the identity attributes are set.