Datadog

Datadog is an observability platform for cloud-scale applications. You can integrate your Render services and databases with Datadog to enable fine-tuned metrics, monitoring, and automated alerting.

Although core Postgres metrics are available from the Render Dashboard, integrating with Datadog can provide more detailed metrics about the Postgres instance host environment. You can also use Datadog as a centralized location for dashboards and automated alerts.

Getting started

Sign up for a Datadog account if you don’t already have one. Then, create or retrieve an API key from your Datadog organization settings page:

Datadog API settings

Make sure to create an API key for your organization. The Datadog integration doesn’t support using an application key or a user-scoped API key.

You can confirm that you’ve correctly generated an API key by calling Datadog’s validate endpoint with your key.

Setting up Postgres monitoring

Adding an API key enables a Datadog agent to run alongside your Postgres instance and report metrics to your Datadog account. All metrics reported by the agent are native to the Datadog platform, so you aren’t billed for custom metrics.

Render currently supports only the US1 Datadog site, which uses the app.datadoghq.com domain. Postgres monitoring is not supported with other Datadog sites (such as EU1, US3, and US5).

For new databases

While creating a Postgres database, provide your Datadog API key in the corresponding field:

Add Datadog for new databases

For existing databases

Add your Datadog API key from the Info tab of your database’s page on the Render Dashboard (in the General section, click Add Datadog API Key).

This requires a restart of your Postgres instance, which causes brief downtime.

Add Datadog for existing databases

Available metrics

Render fully supports all of the following Datadog integrations (see the linked documentation for metrics details):

IntegrationDescription
PostgresMetrics related to your PostgreSQL instance
DiskMetrics related to disk usage and IO for your Postgres volume
NetworkMetrics related to TCP/IP network stats of instance

In addition, Render reports the following metrics:

MetricDescription
system.cpu.num_coresThe number of CPUs, as chosen by your database instance type
system.cpu.systemThe percentage of time the CPU spent running the kernel
system.cpu.userThe percentage of time the CPU spent running user space processes
system.mem.freeThe amount of free RAM
system.mem.totalThe total amount of physical RAM, as chosen by your database instance type
system.mem.usedThe amount of RAM in use

Viewing metrics in Datadog

You can view reported metrics from any Datadog dashboard or metrics explorer page. You can filter metrics by the database-id tag equal to your Render Postgres database ID.

Datadog Postgres metrics

Setting up log streams

Currently, Render only supports TCP log forwarding with TLS. Check the Datadog docs to confirm whether TCP log forwarding is supported for your site.

You can use Datadog as a logging provider for Render log streams. This enables you to view and manage logs from any Render service or databases directly from your Datadog dashboard.

To set this up, follow the instructions from the log stream docs, using your Datadog API key and the endpoint that corresponds to your Datadog site:

Datadog SiteEndpoint
US1intake.logs.datadoghq.com:10516
EUtcp-intake.logs.datadoghq.eu:443

Viewing slow query logs

Queries that take longer than 2 seconds are logged with a line that starts with duration: followed by the SQL statement. Here’s an example:

PostgreSQL slow query log