HA crashes and reboots

[ldkud50]

When this integration is enabled and printing eventually HA will crash and reboot. Problem is the error or cause doesn’t persist after reboot, so I don’t know what specifically is causing it. I have excluded all entities from recorder, but it still is happening. If this integration is disabled, no crashes or reboots. HA is a VM on Proxmox with 12gb ram. Anyone else have this issue.

[ldkud50]

I made the following changes and the crash and reboots are gone.

The ws_client.py file has two issues that flood the HA event loop during printing:

_periodic_gets() runs on a 0.2s sleep loop, polling the printer at GET_REQPRINTERPARA_SEC = 5.0s and GET_PRINT_OBJECTS_SEC = 2.0s. This generates a very high volume of state updates during active prints.
on_message is called on every single WebSocket frame with no debouncing — so every printer response fires a full HA state update regardless of whether values actually changed.

When the WebSocket goes dead mid-print and reconnects, a burst of state updates hits HA all at once, which was enough to crash it. The crash wiped the log buffer before any Creality errors could be recorded, making it very hard to diagnose.

Fix:
Three changes to websocket_client.py:

Increase _periodic_gets() sleep from 0.2s → 1.0s
Increase GET_REQPRINTERPARA_SEC from 5.0 → 10.0s and GET_PRINT_OBJECTS_SEC from 2.0 → 10.0s
Add a debounced dispatcher that only fires on_message when values actually changed AND at least 1 second has passed since the last dispatch

python# Add constant
STATE_UPDATE_DEBOUNCE_SEC = 1.0

Add to init

self._last_dispatched_state: dict[str, Any] = {}
self._last_dispatch_time: float = 0.0

Add method

async def _dispatch_if_changed(self, new_state: dict[str, Any]) -> None:
now = time.monotonic()
changed_keys = {
k for k, v in new_state.items()
if self._last_dispatched_state.get(k) != v
}
time_since_last = now - self._last_dispatch_time
if changed_keys and time_since_last >= STATE_UPDATE_DEBOUNCE_SEC:
self._last_dispatched_state = dict(new_state)
self._last_dispatch_time = now
await self._on_message(new_state)

Replace on_message call in _loop with:

await self._dispatch_if_changed(dict(self._state))

Change sleep in _periodic_gets:

await asyncio.sleep(1.0) # was 0.2

Result:
After applying the fix, the WebSocket reconnect event now logs cleanly as K WS connection dead (no RX for 41.4s); reconnecting and HA stays up. Previously this same event was silently crashing HA. Print is currently running with no crashes.

Happy to share the full modified ws_client.py if helpful.

[3dg1luk43]

Hey, could you please submit a PR? Ioverlook this discussion completely. Tho, very strange! My HA runs on 3GB of RAM and single low power core and I never had any issues like this