Connection crash

[burgainjustin-del]

Sometimes the connection to the device crash and i need to do a power off / power on.
I start with HA, maybe there is some log? Where?
Tk for the app
Justin

[corekill]

Hi Justin, this sounds exactly like the pump's session-table exhaustion we spent several releases fighting. Quick background: the pump firmware has a tiny table of client sessions and is extremely slow to reclaim dead ones. If the integration reconnects too eagerly, the table eventually fills up and the pump refuses all new logins until you power-cycle it — which matches your "crash until power off/on" description.

First step: check your version and update to the latest (v2.3.0). The fixes landed incrementally:

• v2.1.0/v2.1.1 — one persistent session with keepalive, conservative reconnect cooldowns (this is the main fix for your symptom),
• v2.2.0 — lost UDP packets are retransmitted instead of failing the poll,
• v2.3.0 — protection-switch decoding (your other issue).

After updating: power-cycle the heat pump once (unplug ~10 s) to clear its session table, then restart Home Assistant.

Where to find logs: Settings → System → Logs (filter for poolcomfort), or the file /config/home-assistant.log. For more detail you can enable debug logging in configuration.yaml:

logger:
  logs:
    custom_components.poolcomfort: debug

Look for lines like Pool Comfort session lost, connect failed, or reconnect cooldown — if it still gets stuck on v2.3.0, paste those lines here (plus how long the pump had been running since its last power-cycle) and I'll dig in.