Datasource

​

Datasource page

1. Search the catalog or filter it by category: Recommended, Infrastructure, Cloud, Databases, Messaging, Apps & languages, Security, and Custom. The signal tabs (Logs, Metrics, Traces, Profiles) narrow the list further.
2. Select a source such as Kubernetes, a Linux host, AWS, an OpenTelemetry SDK, or Web RUM.
3. Follow the three steps of the guide:
   • Endpoint and token shows where to send data and the token to authenticate with. Show token reveals the token, with its role, workspace, and expiry. For a RUM source, choose the RUM application first: its client token appears once it is chosen.
   • Set up the source shows commands and configuration with the real endpoint and token already filled in.
   • Verify data is arriving checks the receiver, then looks for data received in the last hour. Verify connection checks at once, and the figures refresh every 30 seconds.
4. Read the result of the verification:
   • Data detected: data arrived in the last hour, and the signals of the source are healthy. View received data opens the data, and Create pipeline starts a pipeline for it.
   • Receiver is healthy, but data needs attention: the receiver answers, but no data arrived, or not every signal is healthy. Check the deployment, or wait for the next batch.
   • Verification failed: the receiver could not be checked. Possible causes lists a collector that is not running, a token that is invalid or expired, and a network or proxy that blocks the endpoint.

​

Intake tokens

​

Endpoints and protocols

​

Limits that affect senders

• Request size. A request body larger than 2 MiB is rejected with 413. Keep batches below that size.
• Compression. OTLP over HTTP does not accept compressed request bodies and answers 400. Disable compression in the sender; the OpenTelemetry Collector otlphttp exporter compresses with gzip by default and needs compression: none. Elasticsearch _bulk, Loki push, and the push connectors accept Content-Encoding: gzip.
• Loki encoding. Only the JSON request shape is decoded. Promtail and Grafana Alloy send snappy-compressed protobuf by default and cannot send to this endpoint.
• Rate limit. The router role limits requests per organization and answers 429 with a Retry-After header. A standalone node does not apply this limit. See Role-based deployment.
• Reserved stream. The stream name _molesignal is reserved for the service’s own telemetry and rejects public intake.

​

Streams and schema

​

Native HTTP JSON

curl -X POST http://localhost:5080/api/v1/intake/logs/app \
  -H "authorization: Bearer $MS_JWT" \
  -H 'content-type: application/json' \
  -d "[{\"_timestamp\":$(date +%s)000000,\"level\":\"error\",\"msg\":\"db pool exhausted\",\"trace_id\":\"abc123\"}]"

{"accepted":1,"rejected":0,"errors":[]}

​

OpenTelemetry

export OTEL_EXPORTER_OTLP_PROTOCOL=http/protobuf
export OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:5080/api
export OTEL_EXPORTER_OTLP_HEADERS="authorization=Bearer ${MS_TOKEN}"

exporters:
  otlphttp/molesignal:
    endpoint: http://localhost:5080/api
    compression: none
    headers:
      authorization: "Bearer ${env:MS_TOKEN}"

export OTEL_EXPORTER_OTLP_PROTOCOL=grpc
export OTEL_EXPORTER_OTLP_ENDPOINT=http://localhost:4317
export OTEL_EXPORTER_OTLP_HEADERS="authorization=Bearer ${MS_TOKEN}"

exporters:
  otlp/molesignal:
    endpoint: localhost:4317
    tls:
      insecure: true
    headers:
      authorization: "Bearer ${env:MS_TOKEN}"

​

Prometheus remote_write

remote_write:
  - url: http://localhost:5080/api/v1/prometheus/api/v1/write
    authorization:
      credentials: "${MS_TOKEN}"

​

Elasticsearch bulk

curl -X POST http://localhost:5080/api/v1/_bulk \
  -H "authorization: Bearer $MS_JWT" \
  -H 'content-type: application/x-ndjson' \
  --data-binary $'{"index":{"_index":"app"}}\n{"level":"info","msg":"hello"}\n'

​

Loki push

{"streams":[{"stream":{"service":"checkout","level":"error"},"values":[["1700000000000000000","payment failed"]]}]}

​

Related pages