diff --git a/README.md b/README.md
index 99834bd..f7f1464 100644
--- a/README.md
+++ b/README.md
@@ -28,6 +28,22 @@ InvenTree imports so it is unit-tested here:
 | any other fault | `ERROR`, carrying the reason |
 | agent will / shutdown | `DISCONNECTED` |
 
+### When it refreshes
+
+The status is read from the retained topic in three places, which between them keep the
+page honest without polling the printer itself:
+
+| when | hook |
+|---|---|
+| machine initialisation | `init_machine` |
+| a manual restart from the Admin Center | `restart_machine` |
+| periodically | `ping_machines`, driven by InvenTree's `MACHINE_PING_ENABLED` (on by default) |
+
+The periodic one is not a nicety. Machine status lives in the Django cache, so clearing
+that cache drops every `machine:*` key at once — and this deployment runs Redis without
+persistence, so a restart does exactly that. With only `init_machine` the page then sat
+at `UNKNOWN` with an empty status text until someone printed or the pods restarted.
+
 Two deliberate distinctions. **Unreported media is not healthy media**: `media_ok` is a
 tri-state, and a printer that has said nothing shows "media unreported" rather than
 "media ok". And **battery percentage alone is not enough** — it pins at 100% whenever
diff --git a/inventree_label_dispatch/driver.py b/inventree_label_dispatch/driver.py
index 892e2da..00305f5 100644
--- a/inventree_label_dispatch/driver.py
+++ b/inventree_label_dispatch/driver.py
@@ -110,6 +110,37 @@ def _apply(self, machine, member: str, text: str) -> None:
         machine.set_status(getattr(LabelPrinterStatus, member))
         machine.set_status_text(text)
 
+    def restart_machine(self, machine) -> None:
+        """Re-read the printer's state after a manual restart.
+
+        A restart clears the machine's shared state, and without this nothing puts it
+        back: the page then shows UNKNOWN with an empty status text -- the defaults
+        `get_shared_state` returns for missing keys -- while the machine itself
+        reports `initialized`, no errors and a driver available. An empty status text
+        is diagnostic on its own, since no path in this driver produces one.
+        """
+        self._refresh_status(machine)
+
+    def ping_machines(self) -> None:
+        """Periodic refresh, driven by InvenTree's MACHINE_PING_ENABLED (default on).
+
+        This is what keeps the page honest between prints. It matters more than it
+        looks: machine status lives in the Django cache, so when that cache is cleared
+        -- a Redis restart, and this deployment runs Redis deliberately without
+        persistence -- every `machine:*` key vanishes at once. Before this hook the
+        only thing that re-read the retained topic was machine initialisation, so the
+        page stayed blank until someone printed or the pods restarted.
+
+        One machine's broker trouble must not stop the others being refreshed, hence
+        the per-machine guard. `_refresh_status` already turns an unreachable broker
+        into an UNKNOWN status rather than an exception; this catches anything else.
+        """
+        for machine in self.get_machines():
+            try:
+                self._refresh_status(machine)
+            except Exception as exc:  # noqa: BLE001 - one machine must not break the loop
+                machine.handle_error(exc)
+
     def _refresh_status(self, machine) -> None:
         printer_id = machine.get_setting("PRINTER_ID", "D")
         try:
diff --git a/inventree_label_dispatch/version.py b/inventree_label_dispatch/version.py
index 91ce6a5..77d41e4 100644
--- a/inventree_label_dispatch/version.py
+++ b/inventree_label_dispatch/version.py
@@ -1,3 +1,3 @@
 """Single source of truth for the plugin version."""
 
-__version__ = "0.2.0"
+__version__ = "0.2.1"