After successfully registering vSphere Kubernetes Service (VKS) clusters within a shared Argo CD Supervisor Service instance, the Argo CD Web UI displays a connection status of Unknown for the newly added clusters under the Settings > Clusters view.
This behavior can occur even if the cluster registration command succeeded without errors, and regardless of the underlying infrastructure configuration
vSphere Kubernetes Services
This behavior is by design within upstream Argo CD and does not indicate an infrastructure, networking, or certificate failure.
To optimize performance and minimize unnecessary CPU, memory, and network overhead, the argocd-application-controller uses a lazy-monitoring mechanism. The controller will not actively poll, handshake, or monitor the API server endpoint of a target cluster unless there is at least one active Argo CD Application resource explicitly targeted to deploy workloads onto that specific cluster.
Until an application is assigned, Argo CD leaves the connection status in an idle Unknown state
This is expected behavior and can be ignored, In order to validate that there is no underlying issue deploy a lightweight test or dummy application targeted to these and the application should flip to healthy