The documentation of how to remove a managed cluster from ZTP GitOps pipeline states that in some occasions the cluster may get stuck in the detach process. In that case manual intervention to delete a cluster from ACM is required.

Source: https://docs.openshift.com/container-platform/4.13/scalability_and_performance/ztp_far_edge/ztp-deploying-far-edge-sites.html#ztp-site-cleanup_ztp-deploying-far-edge-sites

This behaviour breaks the GitOps way of working and can have a potential  impact in production scenarios when decommissioning multiple clusters. 

We suggest to remove that note from the documentation and enforce a full GitOps approach. If a cluster gets stuck in the detaching step, there should be a bug report for it and  finally a fix to improve overall product quality.

Not clear how that manual cleanup is required in current versions

1.
2.
3.

If a cluster gets stuck in detach state, post-cleanup is required.

A cluster should not get stuck in detach state when removing it from GitOps pipeline.

System impact: Low/medium. Production environments with a high number of sites may require manual cleanup.

https://docs.openshift.com/container-platform/4.13/scalability_and_performance/ztp_far_edge/ztp-deploying-far-edge-sites.html#ztp-site-cleanup_ztp-deploying-far-edge-sites