Pre and Post Delete Checks
Sveltos provides a robust framework to ensure your managed clusters are in the correct state throughout the entire lifecycle of a resource—from initial deployment to final deletion. This is achieved through three types of checks: ValidateHealthChecks, PreDeleteChecks, and PostDeleteChecks.
These checks are defined within the spec section of a ClusterProfile or a Profile.
Validate Health Checks
ValidateHealthChecks are executed after resources are deployed. They ensure that the application or infrastructure is not just present, but actually functional.
For more details on how to configure these during deployment, see Depends On with Health Checks.
Pre-Delete Checks
PreDeleteChecks are executed before Sveltos begins the deletion process. If any of these checks fail, Sveltos will halt the deletion. This is essential for ensuring that prerequisites (like data backups or draining connections) are met before infrastructure is torn down.
spec:
preDeleteChecks:
- name: "check-no-active-pods"
featureID: Resources
group: ""
version: "v1"
kind: "Pod"
namespace: "production"
script: |
function evaluate()
-- If any pods are found, we return health=false to stop deletion
return { health = false, message = "Active pods still exist in production" }
end
Post-Delete Checks
PostDeleteChecks are executed after Sveltos has issued the delete commands for all resources. These checks ensure the environment has reached a "clean" state, verifying that no lingering resources (like orphaned LoadBalancers or PVCs) remain.