Multi-receiver configuration for PostgreSQL with NRDOT

You can configure the NRDOT Collector to monitor multiple PostgreSQL instances from one collector. This is useful if you have multiple PostgreSQL databases running on different hosts or ports and want to collect metrics from all of them using a single NRDOT Collector instance.

Prerequisites

NRDOT is installed and configured for your PostgreSQL instances:
- Self-hosted PostgreSQL
- AWS RDS/Aurora

Add multi-receiver configuration

Select the deployment type that matches your environment. You can add more receivers as needed for additional PostgreSQL instances.

The following example shows how to configure the NRDOT Collector to monitor two self-hosted PostgreSQL instances using a multi-receiver configuration. The first instance, nrpostgresql/db1, carries the complete configuration and acts as the anchor; the second instance, nrpostgresql/db2, inherits every shared setting through the YAML merge key and overrides only its own connection details.

receivers:
  # Base PostgreSQL instance configuration (Anchor; subsequent instances inherit via merge key)
  nrpostgresql/db1: &nrpostgresql-common
    endpoint: "<YOUR_DB1_HOST>:<YOUR_DB1_PORT>"
    username: "<YOUR_DB_USERNAME>"
    password: "<YOUR_DB_PASSWORD>"
    databases:
      - <YOUR_DB1_NAME>
    collection_interval: 15s

    events:
      db.server.top_query:
        enabled: true
      db.server.query_sample:
        enabled: true
      db.server.query_plan:
        enabled: true

    top_query_collection:
      max_rows_per_query: 1000
      top_n_query: 200
      collection_interval: 60s
      allowed_comment_keys: [nr_service_guid]

    query_sample_collection:
      max_rows_per_query: 1000
      allowed_comment_keys: [nr_service_guid]

    resource_attributes:
      db.system.version:
        enabled: true

    metrics:
      postgresql.database.locks:
        enabled: true
      postgresql.deadlocks:
        enabled: true
      postgresql.function.calls:
        enabled: true
      postgresql.query.conflicts:
        enabled: true
      postgresql.sequential_scans:
        enabled: true
      postgresql.temp.io:
        enabled: true
      postgresql.temp_files:
        enabled: true

  # Second PostgreSQL instance configuration
  nrpostgresql/db2:
    <<: *nrpostgresql-common
    endpoint: "<YOUR_DB2_HOST>:<YOUR_DB2_PORT>"
    username: "<YOUR_DB_USERNAME>"
    password: "<YOUR_DB_PASSWORD>"
    databases:
      - <YOUR_DB2_NAME>

processors:
  batch:
  batch/metrics:
    send_batch_size: 8192
    send_batch_max_size: 8192
    timeout: 10s

exporters:
  otlp/newrelic:
    endpoint: "<YOUR_NEWRELIC_OTLP_ENDPOINT>"
    headers:
      api-key: "<YOUR_NEWRELIC_LICENSE_KEY>"
    compression: gzip
    sending_queue:
      enabled: true
      sizer: bytes
      queue_size: 100_000_000
      num_consumers: 10
      batch:
        sizer: bytes
        max_size: 1_000_000
        min_size: 0
        flush_timeout: 5s
    retry_on_failure:
      enabled: true
      initial_interval: 5s
      max_interval: 30s
      max_elapsed_time: 300s

service:
  telemetry:
    metrics:
      level: none
  pipelines:
    metrics:
      receivers: [nrpostgresql/db1, nrpostgresql/db2]
      processors: [batch/metrics]
      exporters: [otlp/newrelic]
    logs:
      receivers: [nrpostgresql/db1, nrpostgresql/db2]
      processors: [batch]
      exporters: [otlp/newrelic]

The <<: *nrpostgresql-common merge key inherits all settings from nrpostgresql/db1, including its events, query collection settings, and metrics overrides. Any field you explicitly set after it, such as endpoint and databases, overrides the inherited value for that specific receiver instance.

The following example shows how to configure the NRDOT Collector to monitor two RDS or Aurora PostgreSQL instances using a multi-receiver configuration. The first instance, nrpostgresql/db1, carries the complete configuration, including exclude_databases for reserved databases, and acts as the anchor; the second instance, nrpostgresql/db2, inherits every shared setting through the YAML merge key and overrides only its own connection details.

receivers:
  # Base PostgreSQL instance configuration (Anchor; subsequent instances inherit via merge key)
  nrpostgresql/db1: &nrpostgresql-common
    endpoint: "<YOUR_DB1_HOST>:<YOUR_DB1_PORT>"
    username: "<YOUR_DB_USERNAME>"
    password: "<YOUR_DB_PASSWORD>"
    databases:
      - <YOUR_DB1_NAME>
    # top_query/query_sample scan pg_stat_statements/pg_stat_activity cluster-wide
    # regardless of the databases list above; on RDS/Aurora, exclude reserved
    # databases the monitoring user can never reach.
    exclude_databases:
      - rdsadmin
    collection_interval: 15s

    events:
      db.server.top_query:
        enabled: true
      db.server.query_sample:
        enabled: true
      db.server.query_plan:
        enabled: true

    top_query_collection:
      max_rows_per_query: 1000
      top_n_query: 200
      collection_interval: 60s
      allowed_comment_keys: [nr_service_guid]

    query_sample_collection:
      max_rows_per_query: 1000
      allowed_comment_keys: [nr_service_guid]

    resource_attributes:
      db.system.version:
        enabled: true

    metrics:
      postgresql.database.locks:
        enabled: true
      postgresql.deadlocks:
        enabled: true
      postgresql.function.calls:
        enabled: true
      postgresql.query.conflicts:
        enabled: true
      postgresql.sequential_scans:
        enabled: true
      postgresql.temp.io:
        enabled: true
      postgresql.temp_files:
        enabled: true

  # Second PostgreSQL instance configuration
  nrpostgresql/db2:
    <<: *nrpostgresql-common
    endpoint: "<YOUR_DB2_HOST>:<YOUR_DB2_PORT>"
    username: "<YOUR_DB_USERNAME>"
    password: "<YOUR_DB_PASSWORD>"
    databases:
      - <YOUR_DB2_NAME>

processors:
  batch:
  batch/metrics:
    send_batch_size: 8192
    send_batch_max_size: 8192
    timeout: 10s

exporters:
  otlp/newrelic:
    endpoint: "<YOUR_NEWRELIC_OTLP_ENDPOINT>"
    headers:
      api-key: "<YOUR_NEWRELIC_LICENSE_KEY>"
    compression: gzip
    sending_queue:
      enabled: true
      sizer: bytes
      queue_size: 100_000_000
      num_consumers: 10
      batch:
        sizer: bytes
        max_size: 1_000_000
        min_size: 0
        flush_timeout: 5s
    retry_on_failure:
      enabled: true
      initial_interval: 5s
      max_interval: 30s
      max_elapsed_time: 300s

service:
  telemetry:
    metrics:
      level: none
  pipelines:
    metrics:
      receivers: [nrpostgresql/db1, nrpostgresql/db2]
      processors: [batch/metrics]
      exporters: [otlp/newrelic]
    logs:
      receivers: [nrpostgresql/db1, nrpostgresql/db2]
      processors: [batch]
      exporters: [otlp/newrelic]

The <<: *nrpostgresql-common merge key inherits all settings from nrpostgresql/db1, including its exclude_databases list, events, query collection settings, and metrics overrides. Any field you explicitly set after it, such as endpoint and databases, overrides the inherited value for that specific receiver instance.

Restart NRDOT Collector

Restart the NRDOT Collector service for configuration changes to take effect. The collector loads configuration parameters once at startup and doesn't support hot-reloading. Any changes made to the configuration while the collector is running remain inactive until the process restarts.

bash

$sudo systemctl restart nrdot-collector

Compatibility and prerequisites

Learn about supported PostgreSQL versions, required grants, and network requirements.

Metrics reference

Learn about the available metrics collected by the NRDOT Collector.

Troubleshooting guide

Learn how to troubleshoot common issues with PostgreSQL monitoring.