The remediation strategy is the way to tell Burrito how it should handle the remediation of drifts on your Terraform layers.
As for the runner spec override, you can specify a spec.remediationStrategy either on the TerraformRepository or the TerraformLayer.
The configuration of the TerraformLayer will take precedence.
| Field | Type | Default | Effect |
|---|---|---|---|
autoApply |
Boolean | false |
If true when a plan shows drift, it will run an apply. |
nonDestructiveApply |
Boolean | false |
If true, apply runs are skipped when the latest plan includes resources to destroy. |
onError.maxRetries |
Integer | 5 or value defined in Burrito configuration |
How many times Burrito should retry a plan/apply when a runner fails. |
!!! warning
This operator is still experimental. Use spec.remediationStrategy.autoApply: true at your own risk.
!!! note
nonDestructiveApply only applies when autoApply is enabled. If the latest plan summary includes resources to delete, Burrito will keep the layer in ApplyNeeded and wait for the next drift detection instead of creating an apply run.
With this example configuration, Burrito will create apply runs for this layer, with a maximum of 3 retries.
apiVersion: config.terraform.padok.cloud/v1alpha1
kind: TerraformLayer
metadata:
name: random-pets-terragrunt
spec:
remediationStrategy:
autoApply: true
nonDestructiveApply: true
onError:
maxRetries: 3
# ... snipped ...