EDOT running in Gateway mode - "sending queue is full"

Elastic Stack Elastic Agent
1

When using recommended elastic/elastic-agent image running in OTEL Gateway mode it shows errors when ingesting 10m and 60m aggregated data.

Example setup

  1. Start Elastic server using elasticsearch/elasticsearch:9.2.1 Docker image.

  2. Generate API key to use for agent

  3. Start elastic agent image elastic/elastic-agent:9.2.1 in EDOT Gateway mode ( Default configuration of the EDOT Collector (standalone) | Elastic Agent ) using recommended defaults from the same page (can't post github link here)

  4. Start some simple app and point it towards Gateway, example:


// my-app.js

import express from 'express';

const app = express();

const port = process.env.PORT || 3000;

app.get('/hello', (_req, res) => {

res.type('text/plain').send('Hello, World!');

});

if (process.env.NODE_ENV !== 'test') {

app.listen(port, () => {

console.log(`my-app listening on http://localhost:${port}`);

});

}

export default app;



export OTEL_EXPORTER_OTLP_ENDPOINT="<GATEWAY_URL>"

export OTEL_EXPORTER_OTLP_PROTOCOL="grpc"

export OTEL_RESOURCE_ATTRIBUTES="deployment.environment.name=dev,service.name=sample-app"

node --import ` @elastic`/opentelemetry-node my-app.js
  1. Check EDOT Gateway logs, after a while when 10m or 60m metrics are supposed to be exported, you’ll see following error message:

2026-01-20T07:40:00.001Z error internal/base_exporter.go:114 Exporting failed. Rejecting data. {"resource": {"service.instance.id": "dd20735c-d1d9-4976-84da-e07a145c8ee6", "service.name": "elastic-agent", "service.version": "9.2.3"}, "otelcol.component.id": "debug", "otelcol.component.kind": "exporter", "otelcol.signal": "metrics", "error": "sending queue is full", "rejected_items": 8}

go.opentelemetry.io/collector/exporter/exporterhelper/internal.(*BaseExporter).Send

go.opentelemetry.io/collector/exporter/exporterhelper@v0.139.0/internal/base_exporter.go:114

go.opentelemetry.io/collector/exporter/exporterhelper/internal.NewMetricsRequest.newConsumeMetrics.func1

go.opentelemetry.io/collector/exporter/exporterhelper@v0.139.0/internal/new_request.go:176

go.opentelemetry.io/collector/consumer.ConsumeMetricsFunc.ConsumeMetrics

go.opentelemetry.io/collector/consumer@v1.45.0/metrics.go:27

go.opentelemetry.io/collector/service/internal/refconsumer.refMetrics.ConsumeMetrics

go.opentelemetry.io/collector/service@v0.139.0/internal/refconsumer/metrics.go:29

go.opentelemetry.io/collector/internal/fanoutconsumer.(*metricsConsumer).ConsumeMetrics

go.opentelemetry.io/collector/internal/fanoutconsumer@v0.139.0/metrics.go:71

go.opentelemetry.io/collector/consumer.ConsumeMetricsFunc.ConsumeMetrics

go.opentelemetry.io/collector/consumer@v1.45.0/metrics.go:27

github.com/elastic/opentelemetry-collector-components/processor/lsmintervalprocessor.(*Processor).exportForInterval

github.com/elastic/opentelemetry-collector-components/processor/lsmintervalprocessor@v0.20.0/processor.go:597

github.com/elastic/opentelemetry-collector-components/processor/lsmintervalprocessor.(*Processor).export

github.com/elastic/opentelemetry-collector-components/processor/lsmintervalprocessor@v0.20.0/processor.go:463

github.com/elastic/opentelemetry-collector-components/processor/lsmintervalprocessor.(*Processor).commitAndExport

github.com/elastic/opentelemetry-collector-components/processor/lsmintervalprocessor@v0.20.0/processor.go:445

github.com/elastic/opentelemetry-collector-components/processor/lsmintervalprocessor.(*Processor).Start.func1

github.com/elastic/opentelemetry-collector-components/processor/lsmintervalprocessor@v0.20.0/processor.go:201

2026-01-20T07:40:00.001Z warn lsmintervalprocessor@v0.20.0/processor.go:202 failed to export {"resource": {"service.instance.id": "dd20735c-d1d9-4976-84da-e07a145c8ee6", "service.name": "elastic-agent", "service.version": "9.2.3"}, "otelcol.component.id": "elasticapm", "otelcol.component.kind": "connector", "otelcol.signal": "traces", "otelcol.signal.output": "metrics", "error": "failed to export: failed to export interval 10m0s for end time 1768894800: failed to consume the decoded value: sending queue is full", "end_time": "2026-01-20T07:40:00.000Z"}