Datadog
Sends processed data to Datadog Logs. Supports streaming of security events and logs for monitoring, analysis, and alerting.
Requirements
To use the Datadog output connector, you need:
- API Key: A Datadog API key for authentication
- Domain URL: The base domain of your Datadog instance (e.g., datadoghq.com, us5.datadoghq.com)
Obtaining Datadog Credentials
-
Finding Your Domain URL
Your Datadog domain depends on your account region:
datadoghq.com- US1 (default)us3.datadoghq.com- US3us5.datadoghq.com- US5datadoghq.eu- EU1ap1.datadoghq.com- AP1ap2.datadoghq.com- AP2ddog-gov.com- US1-FED
You can verify your region by checking the URL when logged into Datadog (e.g.,
https://app.datadoghq.comorhttps://us5.datadoghq.com). -
Generating an API Key
- Log in to your Datadog account
- Navigate to API Keys
- Click New Key to create a new API key
- Give it a descriptive name (e.g., "Monad Integration")
- Copy and securely store the API key
Details
The Datadog output connector continuously sends processed data to Datadog Logs via the HTTP intake API. The functionality includes the following key features:
-
HTTP Logs Intake: Data is sent to Datadog's HTTP logs intake endpoint at
https://http-intake.logs.<DOMAIN_URL>/api/v2/logs. This ensures reliable delivery of log data to your Datadog account. -
Batch Processing: Logs are processed and sent to Datadog in batches to ensure efficient handling of large volumes of data and optimize API usage.
-
Metadata Enrichment: The connector supports adding metadata to your logs including:
- Source: Identifies the technology from which the log originated. When it matches a Datadog integration name, corresponding parsers and facets are automatically installed
- Service: Links logs to APM services for unified observability
- Hostname: Identifies the originating host
- Tags: Custom tags for filtering and analysis
-
Timestamp Handling: The connector makes sure every record carries a timestamp Datadog can read, converting epoch seconds to epoch milliseconds when needed. See Timestamps below.
-
Error Handling: The connector includes mechanisms to handle errors gracefully, such as logging issues and retrying failed operations to maintain data integrity.
Example data format:
Logs are sent to Datadog in JSON format. Each record is structured as a JSON object with your log data plus any configured metadata:
Code
Timestamps
Datadog parses the JSON in message and uses a field such as timestamp as the log's official date. Datadog reads numeric timestamps as epoch milliseconds only. A record with an epoch-seconds value, for example "timestamp": 1791288415, is read as a date in January 1970.
Logs with epoch-seconds timestamps are silently dropped
Datadog only accepts logs dated within a limited window around the time of ingestion. A log dated in 1970 falls outside that window, so Datadog accepts the request (HTTP 202) but never indexes the log. It does not appear in Log Explorer or Live Tail, and the pipeline reports no errors.
The connector handles this for you in one of two ways, depending on whether Timestamp Field is set.
Timestamp Field set
The connector reads the value at that path from each record and converts it to epoch milliseconds when needed. Use this when your event time is in a field Datadog doesn't check by default (for example created_at or event.time).
- If the field is missing from a record, or its value can't be parsed, that record fails with an error and isn't sent. The other records in the batch are still sent.
Timestamp Field empty
The connector looks for the attributes Datadog's date remapper reads by default, in Datadog's order: timestamp, date, _timestamp, Timestamp, eventTime, published_date. Only top-level attributes are checked.
- Found: the first one present is converted to epoch milliseconds when needed.
- Not found: a
timestampattribute is added with the current time, so the log is dated when Monad sent it. - Found but unparseable: the record is sent unchanged.
When records are given the current time or have an unparseable timestamp, the connector logs a warning to your organization logs. To avoid flooding your logs, it logs at most once every 5 seconds, with the totals since the previous warning.
Conversion rules
- Epoch seconds, epoch milliseconds (including fractional values) and common date-time strings are supported.
- Values Datadog already reads correctly (epoch milliseconds and RFC3339 strings) are left as they are. Other values are rewritten in place as epoch milliseconds.
- The rest of the record is unchanged.
For example, this record:
Code
is sent to Datadog as:
Code
Even with correct timestamps, Datadog drops logs dated too far in the past or future. Backfilled historical data may not appear. See Datadog's custom log forwarding limits.
Configuration
The following configuration defines the output parameters. Each field's specifications, such as type, requirements, and descriptions, are detailed below.
Settings
| Setting | Type | Required | Description |
|---|---|---|---|
| Domain URL | string | Yes | The base domain of the Datadog API (e.g., us5.datadoghq.com). Logs are sent to https://http-intake.logs.<DOMAIN_URL>/api/v2/logs. Valid options: datadoghq.com, us3.datadoghq.com, us5.datadoghq.com, datadoghq.eu, ap1.datadoghq.com, ap2.datadoghq.com, ddog-gov.com. Default: datadoghq.com. |
| Source | string | No | The integration name associated with your log: the technology from which the log originated. When it matches an integration name, Datadog automatically installs the corresponding parsers and facets. Examples: nginx, postgresql, aws, kubernetes. |
| Tags | array | No | Tags associated with your logs. Use keyenv:production, team:security). Tags enable filtering and grouping in Datadog. |
| Hostname | string | No | The name of the originating host of the log. This helps identify which server or container generated the log. |
| Service | string | No | The name of the application or service generating the log events. Used to correlate logs with APM data, so use the same value in both Logs and APM products for unified observability. |
| Timestamp Field | string | No | JSON path to the event timestamp in each record (e.g., created_at, $.event.time). The value is converted to epoch milliseconds when needed, which Datadog requires for numeric timestamps. Records where the field is missing or unparseable fail and aren't sent. If left empty, the connector uses the first of Datadog's default date attributes found in the record, or adds timestamp with the current time. See Timestamps. |
Secrets
| Secret | Type | Required | Description |
|---|---|---|---|
| API Key | string | Yes | The API key for authenticating with the Datadog API. Generate this from your Datadog Organization Settings > API Keys. |