Collect Metrics

The Sumo Logic Distribution for OpenTelemetry Collector provides various receivers for metrics collection. The full list of OpenTelemetry receivers are available on OpenTelemetry collector repository.

To use a receiver, define it in the receivers: section, then reference it in a pipeline. Place your configuration files in the conf.d directory.

Host metrics

The following example uses the Host Metrics receiver to collect system-level metrics like CPU, memory, and disk usage:

receivers:
  hostmetrics/linux/localhost:
    scrapers:
      cpu:
        metrics:
          system.cpu.utilization:
            enabled: true
      load: null
      memory:
        metrics:
          system.memory.utilization:
            enabled: true
      disk: null
      filesystem:
        metrics:
          system.filesystem.utilization:
            enabled: true
      network: null
      paging: null
      process:
        mute_process_name_error: true
        metrics:
          process.threads:
            enabled: true
          process.paging.faults:
            enabled: true
          process.cpu.utilization:
            enabled: true
          process.memory.utilization:
            enabled: true
          process.signals_pending:
            enabled: true
          process.open_file_descriptors:
            enabled: true
          process.context_switches:
            enabled: true
          process.disk.operations:
            enabled: true
processors:
  resource/linux_resource_attributes/localhost:
    attributes:
      - key: sumo.datasource
        value: linux
        action: insert
service:
  pipelines:
    metrics/linux/localhost:
      receivers:
        - hostmetrics/linux/localhost
      processors:
        - memory_limiter
        - resource/linux_resource_attributes/localhost
        - resourcedetection/system
        - batch
      exporters:
        - sumologic

Prometheus metrics

If you have applications that expose metrics in Prometheus format, you can scrape them using the prometheus receiver. This is the recommended approach for collecting Prometheus metrics with the Sumo Logic OpenTelemetry Collector.

tip

If you're migrating from a Telegraf-based collection setup, see Collect Prometheus Metrics for the older Telegraf-based approach.

The following example scrapes metrics from an application running on localhost:8080:

receivers:
  prometheus:
    config:
      scrape_configs:
        - job_name: 'my-app-metrics'
          scrape_interval: 10s
          static_configs:
            - targets: ['localhost:8080']
service:
  pipelines:
    metrics:
      receivers: [prometheus]
      exporters: [sumologic]

This configuration:

• Defines a prometheus receiver that scrapes metrics from localhost:8080 every 10 seconds.
• Creates a metrics pipeline that sends the scraped metrics to Sumo Logic via the sumologic exporter.

You can add multiple targets under static_configs to scrape from several endpoints, or add additional scrape_configs entries for different jobs:

receivers:
  prometheus:
    config:
      scrape_configs:
        - job_name: 'app-1'
          scrape_interval: 10s
          static_configs:
            - targets: ['app1-host:9090']
        - job_name: 'app-2'
          scrape_interval: 30s
          static_configs:
            - targets: ['app2-host:9090', 'app2-host:9091']

Once your metrics are flowing into Sumo Logic, you can visualize them in a dashboard and set up monitors to alert on threshold breaches.

note

You'll need to restart the collector process in order to apply changes.

tip

Refer to our Additional Configurations Reference for more details about OpenTelemetry configuration.