Skip to main content
Docs
Feature Requests Onboarding Trust Center Status Page

Community

Connector Improvement: DataDog - Custom Incident Impact Field Ingestion

Answered
Rob Lee User

The Datadog connector currently ingests custom field definitions from the /incident-fields endpoint into the incident_field table, but does not capture the actual per-incident values of custom impact fields.

Custom incident impacts are exposed via the GET /api/v2/incidents/{id}/impacts endpoint. Each impact object contains a nested fields object with custom field values (e.g., fiscal_impact, false_origin_service, incident_cause) that are not currently ingested by the connector.

Request: Add support for ingesting custom incident impact field values from /api/v2/incidents/{id}/impacts and store them in an appropriate table structure (e.g., incident_impact_field).

API Endpoint: GET /api/v2/incidents/{id}/impacts
Documentation: https://docs.datadoghq.com/api/latest/incidents/?code-lang=curl#list-an-incidents-impacts

This would enable downstream analysis of custom impact metrics without requiring custom API calls outside of Fivetran.

Please sign in to leave a comment.

Comments

2 comments

Date Votes
  • Official comment
    Sisan Kumar Sahu Fivetranner

    Hi Rob,

    Thanks for the input and the API link. I see that the API is in Preview and not yet GA - one of the reasons why we have not added support for it yet.  Nevertheless, we have added this to the backlog and will keep you posted as we make progress.

    Regards,
    Sisan

  • Rob Lee User

    Hi Sisan, we just heard back from our support contact at Datadog. It sounds like this endpoint is in fact stable, they just haven't updated the docs. Here's their response:

    Regarding the GET /api/v2/incidents/{incident_id}/impacts endpoint, I also confirmed with the team that this endpoint is stable, and will work with our docs team to update this status in our docs accordingly. Thanks for raising this to our attention!

    So, it sounds like GA is happening soon.

    Thanks,

    Rob