Reference — Metrics

The telemetry-processor emits pipeline throughput metrics through the EdgeCommons metric service. With metricEmission.target: messaging, metrics are published on the reserved UNS metric class:

ecv1/{device}/telemetry-processor/metric/pipeline

The processor does not publish directly to reserved metric topics. It defines and emits the metric through gg.metrics(), so the same metric name, measures, units, and dimensions are used by messaging, CloudWatch, and Prometheus targets.

Dimension model

The pipeline metric is intentionally fleet-level for the processor instance. It uses runtime-injected component dimensions and does not include route ids as CloudWatch dimensions. Per-route details are available through the get-stats command.

This keeps the CloudWatch dimension set stable even when routes are added, removed, or renamed.

Emission model

The processor emits pipeline every 30 seconds. Measures are interval deltas summed across all routes, not lifetime totals. A restarted process never emits negative deltas; counters saturate at zero when a previous snapshot is greater than the current snapshot.

System measures such as CPU and memory are emitted separately by the EdgeCommons runtime as the sys metric. This page describes only the processor’s custom non-system metric.

pipeline

Fleet-level route throughput and dispatch health.

Dimensions: runtime-injected component dimensions only.

Measure	Unit	Purpose
messagesIn	Count	Messages accepted into route worker queues during the interval. Helps measure inbound telemetry load.
messagesOut	Count	Messages successfully forwarded or appended during the interval. Helps measure successful pipeline output.
messagesDropped	Count	Messages dropped because route queues were full. Helps detect backpressure and undersized maxQueue settings.
streamAppends	Count	Records appended to durable stream targets. Helps confirm stream-backed routes are producing output.
publishFailures	Count	Local/northbound publish failures and stream append failures. Helps detect downstream broker, IoT Core, or stream sink problems.

Related command data

Use the get-stats command for per-route counters:

ecv1/{device}/telemetry-processor/cmd/get-stats

The command returns route ids, cumulative in/out/drop counts, stream appends, publish failures, queue depth, and paused state. Use it when you need route-level diagnosis that would be too high-cardinality for CloudWatch metric dimensions.