Skip to main content

Native Datadog and PagerDuty Integration for Alert Silencing

Feature Requests

During deployments, we silence alerts in Datadog to avoid unnecessary noise and downstream PagerDuty incidents.

Currently, we handle this using custom Python scripts in pre-deploy hooks. It works, but it would be much better if Spacelift had a built-in integration for change window alert silencing with Datadog and potentially PagerDuty.

A native integration would:

  • Reduce custom scripting

  • Make deployments cleaner

  • Improve operational safety

This would be especially useful for teams that rely heavily on monitoring integrations.

Status: ❌ Rejected2 comments

Log in to comment and vote

Comments2

  • Coral Chipmunk

    •

    Mar 18

    Hi @Marcin Wyszynski, I kind of assumed that would be response but wanted to ask regardless, thanks.

  • Black Breeze

    •

    Mar 17

    Thanks for the detailed feedback @Rob Put . This is definitely a common operational challenge that many teams face.

    While we understand the value this would bring to your workflow, we've designed Spacelift to focus on infrastructure orchestration rather than expanding into alerting and monitoring management. However, you can achieve this functionality today using Spacelift's existing notification policies and webhook capabilities.

    You can create a notification policy that calls Datadog's downtime API when deployments start. This gives you the same result as a native integration but with full control over the logic and timing.

    Similarly, you can use webhooks to create maintenance windows via PagerDuty's API during your deployment workflows.

    This approach offers several advantages:


    • Works with any monitoring/alerting tool, not just Datadog and PagerDuty
    • Gives you complete control over when and how alerts are silenced
    • Leverages Spacelift's policy engine for sophisticated conditional logic
    • No vendor lock-in or dependency on specific tool integrations

    Would this webhook-based approach work for your use case? If you need more sophisticated logic, we have a new product coming up called Flows where you can declaratively build event-based workflows, which is likely the most suitable way of solving similar problems at scale. Happy to run a demo for you if you want.