Runbook: OTel Collector Down
Recovery steps for the local Nexus OpenTelemetry Collector
Esta página aún no está disponible en tu idioma.
This runbook covers the local Nexus OpenTelemetry Collector used by the dev compose stack.
Symptoms
Section titled “Symptoms”http://127.0.0.1:18888/metricsis unreachable.- Grafana Explore shows no recent Tempo traces or Loki logs.
- Service logs mention OTLP export failures.
Checks
Section titled “Checks”From the repository root:
just compose ps otel-collectorjust logs otel-collectorjust compose config --quietIf the config command fails, fix deploy/compose/otel-collector.yaml before restarting.
Recovery
Section titled “Recovery”just compose up -d otel-collector tempo loki prometheus grafanaThen verify:
curl -fsS http://127.0.0.1:18888/metrics | grep otelcol_receiver_acceptedTier 2 telemetry is operational-only. Do not copy tenant request bodies, API keys, prompts, completions, or tenant IDs into collector logs or debugging payloads.
See Observability (Local Stack) for the full LGTM stack this collector feeds, and Telemetry Tiering for what Tier 2 telemetry may and may not contain.
