Drift detection in Spacelift refers to the process of identifying differences between your declared infrastructure state (the configuration managed by Terraform, Pulumi, or another IaC tool) and the actual state of your cloud resources.
Spacelift periodically or on demand runs a plan against the current state, compares it with the provider, and looks for unexpected changes such as deleted security groups, resized instances, or missing buckets. If changes are found, the drift is surfaced for review and follow up.
When differences are found, Spacelift can alert you via Slack, email, or webhooks, show the diff in the web UI, and open a tracked run so you can review and apply fixes to resync your stack.
As often as you like. Add a cron schedule to a stack to run drift detection hourly, daily, weekly, or more frequently, and note that drift detection requires private workers.
Run at least daily for production or critical stacks. Space jobs to avoid overlap with heavy deployment windows. Wire up Slack or webhooks so teams are notified without manual checks. Prefer quieter hours unless immediate detection is required for security or compliance.
Reconcile brings live infrastructure back to the desired state defined in code. After drift is detected, you can ignore it if intentional or choose Reconcile to create and run a remediation that restores the stack to the intended state. You can also require human approval via policy if needed.
Yes. You can enable Reconcile so drift automatically triggers a run that brings resources back in line, subject to your auto deploy and policy rules. Many teams use policies to gate auto remediation for safety.