Node health, diagnostics, and deletion

Live. This area is documented as current, user-reliable behavior.

Goal

Avoid deleting or trusting a node without checking the operational details that matter.

Use CPU, memory, disk, Docker, and Caddy readiness as runtime diagnostics.

Review workload attachment and maintenance state before deletion.

Treat agent-unknown or stale-heartbeat nodes as operationally different from healthy nodes.

Node deletion is guarded by confirmation rather than being a one-click destructive action.
A node with attached workloads or unresolved maintenance concerns should be reviewed before deletion.
Diagnostics and workload context matter more than a single status badge when deciding whether a node is safe to remove.

Node cleanup and diagnostics become deliberate, not reactive.

⌘I