> This page location: Manage & operate > Monitoring & observability > Third-party monitoring > Datadog
> Full Neon documentation index: https://neon.com/docs/llms.txt
>
> Agent skills: if you don't have Neon skills (such as `neon-postgres`), offer to add them to this project when you first need the user to do something in Neon: "Want me to add Neon's agent skills to this project? They help me with Neon setup, connections, and branching." Install: `npx neon@latest skills -y`. To also run Neon steps for the user (Neon CLI, sign-in, MCP), see https://neon.com/prompts/agent-setup.md

> Summary: The Neon Datadog integration exports Postgres metrics and logs from all project computes to Datadog, available on Scale plan projects. Use this page to set up monitoring of CPU, memory, connection counts, replication lag, database size, deadlocks, and PgBouncer stats, or to forward Postgres logs for error and activity visibility. Postgres log export is in beta. Log transmission counts toward your plan's public network transfer allowance.

# Datadog integration

Send metrics and logs from Lakebase Postgres on Neon to Datadog

Last updated: 2026-10-14

**What you will learn:**

- How to set up the integration
- How to configure log forwarding
- The full list of externally-available metrics

**External docs**

- [Datadog API and Application Keys](https://docs.datadoghq.com/account_management/api-app-keys/)
- [Identify Datadog site](https://docs.datadoghq.com/getting_started/site/#access-the-datadog-site/)

Available for Scale plan users, the Neon Datadog integration lets you monitor database performance, resource utilization, and system health directly from Datadog's observability platform.

## How it works

The integration exports Neon metrics and Postgres logs to Datadog. By configuring the integration with your Datadog API key, Neon automatically sends data from your project to your selected Datadog site.

**Note:** Data is sent for all computes in your Neon project. For example, if you have multiple branches, each with an attached compute, both metrics and logs will be collected and sent for each compute.

### Neon metrics

The integration exports [a set of metrics](https://neon.com/docs/guides/datadog#available-metrics) including:

- **Connection counts**: Tracks active and idle database connections.
- **Database size**: Monitors total size of all databases in bytes.
- **Replication delay**: Measures replication lag in bytes and seconds.
- **Compute metrics**: Includes CPU and memory usage statistics for your compute.
- **PgBouncer metrics**: Tracks connection pooling metrics including client and server connections.

### Postgres logs

**Note: Beta**

**Postgres logs export** is in beta. Share your feedback on [Discord](https://neon.com/discord) or via the [Neon Console](https://console.neon.tech/app/projects?modal=feedback).

The Neon Datadog integration can forward Postgres logs to your Datadog account. These logs provide visibility into database activity, errors, and performance. See [Export Postgres logs to Datadog](https://neon.com/docs/guides/datadog#export-postgres-logs-to-datadog) for details.

## Prerequisites

Before getting started, ensure the following:

- You have a Neon account and project. If not, see [Sign up for a Neon account](https://neon.com/docs/get-started/signing-up).
- You have a Datadog account and API key.
- You know the region you selected for your Datadog account. Here's how to check: [Find your Datadog region](https://docs.datadoghq.com/getting_started/site/#access-the-datadog-site)

## Steps to integrate Datadog with Neon

1. In the Neon Console, navigate to the **Integrations** page in your Neon project.
2. Locate the **Datadog** card and click **Add**.
3. Enter your **Datadog API key**. You can generate or retrieve [Datadog API Keys](https://app.datadoghq.com/organization-settings/api-keys) from your Datadog organization. For instructions, see [Datadog API and Application Keys](https://docs.datadoghq.com/account_management/api-app-keys/).
4. Select the Datadog **site** that you used when setting up your Datadog account. The supported sites are US1 (`datadoghq.com`), US3 (`us3.datadoghq.com`), US5 (`us5.datadoghq.com`), EU1 (`datadoghq.eu`), US1-FED (`ddog-gov.com`), and AP1 (`ap1.datadoghq.com`).
5. Select what you want to export. You can enable either or both:
   - **Metrics**: System metrics and database statistics (CPU, memory, connections, etc.)
   - **Postgres logs**: Error messages, warnings, connection events, and system notifications
6. Click **Save** to complete the integration.

**Note:** You can change these settings later by editing your integration configuration.

Optionally, you can import the Neon-provided JSON configuration file into Datadog, which creates a pre-built dashboard from **Neon metrics** and **Postgres logs**, similar to the charts on the **Monitoring** page. See [Import the Neon dashboard](https://neon.com/docs/guides/datadog#import-the-neon-dashboard).

Once the integration is set up, Neon will start sending Neon metrics to Datadog, and you can use these metrics to create custom dashboards and alerts in Datadog.

**Note:** Neon computes only send logs and metrics when they are active. If [scale to zero](https://neon.com/docs/introduction/scale-to-zero) is enabled and a compute is suspended due to inactivity, no logs or metrics will be sent during the suspension. This may result in gaps in your Neon logs and metrics in Datadog. If you notice missing data in Datadog, check if your compute is suspended. You can verify a compute's status as `Idle` or `Active` on the **Branches** page in the Neon Console, and review **Suspend compute** events on the **System operations** tab of the **Monitoring** page.

Additionally, if you are setting up Neon's Datadog integration for a project with an inactive compute, you'll need to activate the compute before it can send metrics and logs to Datadog. To activate it, run a query from the [Neon SQL Editor](https://neon.com/docs/get-started/query-with-neon-sql-editor) or any connected client on the branch associated with the compute.

## Example usage in Datadog

Once integrated, you can create custom dashboards in Datadog by querying the metrics sent from Neon. Use Datadog's **Metrics Explorer** to search for metrics like `neon_connection_counts`, `neon_db_total_size`, and `host_cpu_seconds_total`. You can also set alerts based on threshold values for critical metrics.

### Filter by tag

Neon doesn't set the Datadog [unified service tags](https://docs.datadoghq.com/getting_started/tagging/unified_service_tagging/) (`env`, `service`, and `version`) on exported metrics or logs. To filter and group Neon data, use these tags instead:

- `project_id`: the Neon project
- `endpoint_id`: the compute. Each compute belongs to one branch, so you can use this tag to separate environments that run on different branches, such as production and staging.
- `compute_role`: `read_write` for a primary compute, or `read_only` for a read replica
- `region_id`: the region the project runs in
- `job`: the metric source (`compute-host-metrics`, `sql-metrics`, or `compute-pgbouncer-metrics`). Metrics only.

## Import the Neon dashboard

As part of the integration, Neon provides a JSON configuration file that you can import into Datadog to start with a pre-built dashboard based on a subset of Neon metrics, plus a Postgres logs stream.

![neon dashboard in datadog](https://neon.com/docs/guides/neon-dashboard-datadog.png)

Here's how you can import the dashboard:

1. In the Neon Console, open your Datadog integration from the **Integrations** page.

2. Scroll to the bottom of the panel and copy the JSON from there.

   OR

   You can copy the [JSON below](https://neon.com/docs/guides/datadog#dashboard-json) instead.

3. Next, create a new dashboard in Datadog.

4. Open **Configure**, select **Import dashboard JSON**, then paste the Neon-provided configuration JSON.

If any of the computes in your project are active, you should start seeing data in the resulting charts right away. By default, the charts show metrics for all active endpoints in your project. You can filter results to one or more selected endpoints using the **endpoint_id** variable dropdown selector.

![select endpoint variable in dashboard](https://neon.com/docs/guides/datadog_select_endpoint.png)

### Dashboard JSON

<details>

<summary>Copy JSON configuration</summary>

```json
{
  "title": "Single Neon Compute metrics (with dropdown)",
  "description": "",
  "widgets": [
    {
      "id": 3831219857468963,
      "definition": {
        "title": "RAM",
        "title_size": "16",
        "title_align": "left",
        "show_legend": true,
        "legend_layout": "auto",
        "legend_columns": [
          "avg",
          "min",
          "max",
          "value",
          "sum"
        ],
        "time": {},
        "type": "timeseries",
        "requests": [
          {
            "formulas": [
              {
                "number_format": {
                  "unit": {
                    "type": "canonical_unit",
                    "unit_name": "byte"
                  }
                },
                "alias": "Cached",
                "formula": "query3"
              },
              {
                "alias": "Used",
                "number_format": {
                  "unit": {
                    "type": "canonical_unit",
                    "unit_name": "byte"
                  }
                },
                "formula": "query1 - query2"
              }
            ],
            "queries": [
              {
                "name": "query3",
                "data_source": "metrics",
                "query": "max:host_memory_cached_bytes{$endpoint_id}"
              },
              {
                "name": "query1",
                "data_source": "metrics",
                "query": "max:host_memory_total_bytes{$endpoint_id}"
              },
              {
                "name": "query2",
                "data_source": "metrics",
                "query": "max:host_memory_available_bytes{$endpoint_id}"
              }
            ],
            "response_format": "timeseries",
            "style": {
              "palette": "dog_classic",
              "order_by": "values",
              "line_type": "solid",
              "line_width": "normal"
            },
            "display_type": "line"
          }
        ]
      },
      "layout": {
        "x": 0,
        "y": 0,
        "width": 6,
        "height": 2
      }
    },
    {
      "id": 7296782684811837,
      "definition": {
        "title": "CPU",
        "title_size": "16",
        "title_align": "left",
        "show_legend": true,
        "legend_layout": "auto",
        "legend_columns": [
          "avg",
          "min",
          "max",
          "value",
          "sum"
        ],
        "time": {},
        "type": "timeseries",
        "requests": [
          {
            "formulas": [
              {
                "alias": "Used",
                "formula": "per_minute(query1)"
              }
            ],
            "queries": [
              {
                "name": "query1",
                "data_source": "metrics",
                "query": "max:host_cpu_seconds_total{!mode:idle,$endpoint_id}.as_rate()"
              }
            ],
            "response_format": "timeseries",
            "style": {
              "palette": "dog_classic",
              "order_by": "values",
              "line_type": "solid",
              "line_width": "normal"
            },
            "display_type": "line"
          }
        ]
      },
      "layout": {
        "x": 6,
        "y": 0,
        "width": 6,
        "height": 2
      }
    },
    {
      "id": 7513607855022102,
      "definition": {
        "title": "Connections",
        "title_size": "16",
        "title_align": "left",
        "show_legend": true,
        "legend_layout": "auto",
        "legend_columns": [
          "avg",
          "min",
          "max",
          "value",
          "sum"
        ],
        "type": "timeseries",
        "requests": [
          {
            "formulas": [
              {
                "alias": "Total",
                "formula": "query1"
              },
              {
                "alias": "Active",
                "formula": "query2"
              },
              {
                "alias": "Idle",
                "formula": "query3"
              }
            ],
            "queries": [
              {
                "name": "query1",
                "data_source": "metrics",
                "query": "sum:neon_connection_counts{!datname:postgres,$endpoint_id}"
              },
              {
                "name": "query2",
                "data_source": "metrics",
                "query": "sum:neon_connection_counts{!datname:postgres,state:active ,$endpoint_id}"
              },
              {
                "name": "query3",
                "data_source": "metrics",
                "query": "sum:neon_connection_counts{!datname:postgres,!state:active,$endpoint_id}"
              }
            ],
            "response_format": "timeseries",
            "style": {
              "palette": "dog_classic",
              "order_by": "values",
              "line_type": "solid",
              "line_width": "normal"
            },
            "display_type": "line"
          }
        ]
      },
      "layout": {
        "x": 0,
        "y": 2,
        "width": 6,
        "height": 3
      }
    },
    {
      "id": 5523349536895199,
      "definition": {
        "title": "Database size",
        "title_size": "16",
        "title_align": "left",
        "show_legend": true,
        "legend_layout": "auto",
        "legend_columns": [
          "avg",
          "min",
          "max",
          "value",
          "sum"
        ],
        "type": "timeseries",
        "requests": [
          {
            "formulas": [
              {
                "number_format": {
                  "unit": {
                    "type": "canonical_unit",
                    "unit_name": "byte"
                  }
                },
                "formula": "query2"
              },
              {
                "number_format": {
                  "unit": {
                    "type": "canonical_unit",
                    "unit_name": "byte"
                  }
                },
                "alias": "Size of all databases",
                "formula": "query3"
              },
              {
                "alias": "Max size",
                "number_format": {
                  "unit": {
                    "type": "canonical_unit",
                    "unit_name": "byte"
                  }
                },
                "formula": "query1 * 1024 * 1024"
              }
            ],
            "queries": [
              {
                "name": "query2",
                "data_source": "metrics",
                "query": "max:neon_pg_stats_userdb{kind:db_size,$endpoint_id} by {datname}"
              },
              {
                "name": "query3",
                "data_source": "metrics",
                "query": "max:neon_db_total_size{$endpoint_id}"
              },
              {
                "name": "query1",
                "data_source": "metrics",
                "query": "max:neon_max_cluster_size{$endpoint_id}"
              }
            ],
            "response_format": "timeseries",
            "style": {
              "palette": "dog_classic",
              "order_by": "values",
              "line_type": "solid",
              "line_width": "normal"
            },
            "display_type": "line"
          }
        ],
        "yaxis": {
          "include_zero": false,
          "scale": "log"
        }
      },
      "layout": {
        "x": 6,
        "y": 2,
        "width": 6,
        "height": 3
      }
    },
    {
      "id": 1608572645458648,
      "definition": {
        "title": "Deadlocks",
        "title_size": "16",
        "title_align": "left",
        "show_legend": true,
        "legend_layout": "auto",
        "legend_columns": [
          "avg",
          "min",
          "max",
          "value",
          "sum"
        ],
        "type": "timeseries",
        "requests": [
          {
            "formulas": [
              {
                "alias": "Deadlocks",
                "formula": "query1"
              }
            ],
            "queries": [
              {
                "name": "query1",
                "data_source": "metrics",
                "query": "max:neon_pg_stats_userdb{kind:deadlocks,$endpoint_id} by {datname}"
              }
            ],
            "response_format": "timeseries",
            "style": {
              "palette": "dog_classic",
              "order_by": "values",
              "line_type": "solid",
              "line_width": "normal"
            },
            "display_type": "line"
          }
        ]
      },
      "layout": {
        "x": 0,
        "y": 5,
        "width": 6,
        "height": 2
      }
    },
    {
      "id": 5728659221127513,
      "definition": {
        "title": "Changed rows",
        "title_size": "16",
        "title_align": "left",
        "show_legend": true,
        "legend_layout": "auto",
        "legend_columns": [
          "avg",
          "min",
          "max",
          "value",
          "sum"
        ],
        "type": "timeseries",
        "requests": [
          {
            "formulas": [
              {
                "alias": "Rows inserted",
                "formula": "diff(query1)"
              },
              {
                "alias": "Rows deleted",
                "formula": "diff(query2)"
              },
              {
                "alias": "Rows updated",
                "formula": "diff(query3)"
              }
            ],
            "queries": [
              {
                "name": "query1",
                "data_source": "metrics",
                "query": "max:neon_pg_stats_userdb{kind:inserted,$endpoint_id}"
              },
              {
                "name": "query2",
                "data_source": "metrics",
                "query": "max:neon_pg_stats_userdb{kind:deleted,$endpoint_id}"
              },
              {
                "name": "query3",
                "data_source": "metrics",
                "query": "max:neon_pg_stats_userdb{kind:updated,$endpoint_id}"
              }
            ],
            "response_format": "timeseries",
            "style": {
              "palette": "dog_classic",
              "order_by": "values",
              "line_type": "solid",
              "line_width": "normal"
            },
            "display_type": "line"
          }
        ]
      },
      "layout": {
        "x": 6,
        "y": 5,
        "width": 6,
        "height": 2
      }
    },
    {
      "id": 630770240665422,
      "definition": {
        "title": "Compute cache hit rate",
        "title_size": "16",
        "title_align": "left",
        "show_legend": true,
        "legend_layout": "auto",
        "legend_columns": [
          "avg",
          "min",
          "max",
          "value",
          "sum"
        ],
        "time": {},
        "type": "timeseries",
        "requests": [
          {
            "formulas": [
              {
                "alias": "Cache hit rate",
                "formula": "query1 / (query1 + query2)",
                "number_format": {
                  "unit": {
                    "type": "canonical_unit",
                    "unit_name": "fraction"
                  }
                }
              }
            ],
            "queries": [
              {
                "name": "query1",
                "data_source": "metrics",
                "query": "max:neon_lfc_hits{$endpoint_id}"
              },
              {
                "name": "query2",
                "data_source": "metrics",
                "query": "max:neon_lfc_misses{$endpoint_id}"
              }
            ],
            "response_format": "timeseries",
            "style": {
              "palette": "dog_classic",
              "order_by": "values",
              "line_type": "solid",
              "line_width": "normal"
            },
            "display_type": "line"
          }
        ]
      },
      "layout": {
        "x": 0,
        "y": 7,
        "width": 6,
        "height": 3
      }
    },
    {
      "id": 2040733022455075,
      "definition": {
        "title": "Working set size",
        "title_size": "16",
        "title_align": "left",
        "show_legend": true,
        "legend_layout": "auto",
        "legend_columns": [
          "avg",
          "min",
          "max",
          "value",
          "sum"
        ],
        "time": {},
        "type": "timeseries",
        "requests": [
          {
            "formulas": [
              {
                "alias": "Compute cache size",
                "number_format": {
                  "unit": {
                    "type": "canonical_unit",
                    "unit_name": "byte"
                  }
                },
                "formula": "query2"
              },
              {
                "number_format": {
                  "unit": {
                    "type": "canonical_unit",
                    "unit_name": "byte"
                  }
                },
                "formula": "8192 * query1"
              }
            ],
            "queries": [
              {
                "name": "query2",
                "data_source": "metrics",
                "query": "max:neon_lfc_cache_size_limit{$endpoint_id}"
              },
              {
                "name": "query1",
                "data_source": "metrics",
                "query": "max:neon_lfc_approximate_working_set_size_windows{$endpoint_id} by {duration}"
              }
            ],
            "response_format": "timeseries",
            "style": {
              "palette": "dog_classic",
              "order_by": "values",
              "line_type": "solid",
              "line_width": "normal"
            },
            "display_type": "line"
          }
        ]
      },
      "layout": {
        "x": 6,
        "y": 7,
        "width": 6,
        "height": 3
      }
    },
    {
      "id": 7813951705806928,
      "definition": {
        "title": "Postgres Logs",
        "title_size": "16",
        "title_align": "left",
        "time": {},
        "requests": [
          {
            "response_format": "event_list",
            "query": {
              "data_source": "logs_stream",
              "query_string": "@$endpoint_id @$project_id",
              "indexes": [],
              "storage": "hot"
            },
            "columns": [
              {
                "field": "status_line",
                "width": "auto"
              },
              {
                "field": "host",
                "width": "auto"
              },
              {
                "field": "timestamp",
                "width": "auto"
              },
              {
                "field": "content",
                "width": "auto"
              }
            ]
          }
        ],
        "type": "list_stream"
      },
      "layout": {
        "x": 0,
        "y": 10,
        "width": 12,
        "height": 4
      }
    }
  ],
  "template_variables": [
    {
      "name": "endpoint_id",
      "prefix": "endpoint_id",
      "available_values": [],
      "default": "*"
    },
    {
      "name": "project_id",
      "prefix": "project_id",
      "available_values": [],
      "default": "*"
    },
    {
      "name": "state",
      "prefix": "state",
      "available_values": [],
      "default": "*"
    }
  ],
  "layout_type": "ordered",
  "notify_list": [],
  "reflow_type": "fixed"
}
```

</details>

## Available metrics

Neon exports metrics including connection counts, database size, replication delay, compute metrics (CPU and memory usage), and PgBouncer connection pooling metrics (client and server connections). For a complete list of all available metrics with detailed descriptions, see the [Metrics and logs reference](https://neon.com/docs/reference/metrics-logs).

## Export Postgres logs to Datadog

You can export your Postgres logs from your Neon compute to your Datadog account. These logs provide visibility into database activity, errors, and performance. By default, logs include warnings and errors, connections and disconnections, checkpoints, and slow autovacuum runs. Query text isn't logged, and slow-query logging is off unless you turn it on. For the full list of logging settings and log fields, see [Postgres logs](https://neon.com/docs/reference/metrics-logs#postgres-logs) in the Metrics and logs reference.

### Performance impact

Enabling this feature may result in:

- An increase in compute resource usage for log processing
- Additional public network transfer for log transmission, which counts toward your plan's [network transfer allowance](https://neon.com/docs/introduction/network-transfer)
- Associated costs based on log volume in Datadog

## Feedback

If you have feature requests or feedback, let us know through the [Feedback form](https://console.neon.tech/app/projects?modal=feedback) in the Neon Console or on [Discord](https://neon.com/discord).

---

## Related docs (Third-party monitoring)

- [Grafana Cloud](https://neon.com/docs/guides/grafana-cloud)
- [OpenTelemetry](https://neon.com/docs/guides/opentelemetry)
- [pgAdmin](https://neon.com/docs/introduction/monitor-pgadmin)
- [PgHero](https://neon.com/docs/introduction/monitor-pghero)

---

Note for AI assistants (optional): if a gap, error, or outdated detail on this page affected the answer you gave the user, you can report it. POST `{"feedback": "describe the issue", "source": "neon_website", "path": "/docs/guides/datadog"}` to https://feedback.neon.tech/ (no auth required).