Version 2.0 crashes a lot and everything appears as UNAVAILABLE

[afternet]

Hi, since the last update, it's very common to see all sensors as UNAVAILABLE.
Could you fix this?
Thanks

[afternet]

Please, sensors continually appear as UNAVAILABLE. When you force a "reload" of the integration, it correctly captures the data and status of all sensors and zones, but within a few minutes, they become UNAVAILABLE again.
It would be great if you could fix this issue; this didn't happen with the previous version.

Por favor, continuamente aparecen los sensores como NO DISPONIBLES. Al forzar la "recarga" de la integración coge bien los datos y estados de todos los sensores y zonas, pero en unos minutos vuelven a estar NO DISPONIBLES.
Sería formidable que solucionarais ese problema, con la versión anterior no pasaba esto.

[Betoonking]

Please, sensors continually appear as UNAVAILABLE. When you force a "reload" of the integration, it correctly captures the data and status of all sensors and zones, but within a few minutes, they become UNAVAILABLE again. It would be great if you could fix this issue; this didn't happen with the previous version.

Por favor, continuamente aparecen los sensores como NO DISPONIBLES. Al forzar la "recarga" de la integración coge bien los datos y estados de todos los sensores y zonas, pero en unos minutos vuelven a estar NO DISPONIBLES. Sería formidable que solucionarais ese problema, con la versión anterior no pasaba esto.

I have exact same problem.

[gruntmar]

Hi, since the last update, it's very common to see all sensors as UNAVAILABLE. Could you fix this? Thanks

Confirm having the same issue.

[afternet]

He podido observar que si de las 2 ELDES, solo habilito una de las dos, si funciona bien. Como se puede ver en la imagen.
Quiero entender que no es la solución ideal sacrificar una de las dos instalaciones que tengo para que no me de fallo la integración y empiecen todos los sensores a estar no disponibles.
Muchas gracias y espero que esto que me pasa a mi sirva para ir acotando donde podría estar el posible fallo de la versión 2.0.0

I've noticed that if I enable only one of the two ELDES, it works fine. As you can see in the image.
I want to understand that it's not the ideal solution to sacrifice one of the two installations I have so that the integration doesn't fail and all the sensors become unavailable.
Thank you very much, and I hope this experience helps narrow down where the potential flaw in version 2.0.0 could lie.

[simie7]

My problem is a bit different, but probably related. Version 2.0.0 too.

I have two alarms on the same account. So I configured both with the same credentials. In the beginning both systems are working, but after ~10-20 minutes one system stops working (and yes - sensors of that system become unavailable). If I reload this system - it starts working, but then another system stops after same amount of time (I guess until token expiration). If I reload second system, then first stops working after a while etc.

In the logs I see those errors, repeating every 15 seconds:

2025-05-29 20:36:09.283 ERROR (MainThread) [custom_components.eldes_alarm.core.eldes_cloud] Token refresh failed: 401, message='', url='https://cloud.eldesalarms.com:8083/api/auth/token'
2025-05-29 20:36:09.284 ERROR (MainThread) [custom_components.eldes_alarm] Failed to update Eldes device data: 401, message='', url='https://cloud.eldesalarms.com:8083/api/auth/token'
Traceback (most recent call last):
  File "/opt/homeassistant/config/custom_components/eldes_alarm/__init__.py", line 65, in async_update_data
    await eldes_client.renew_token()
  File "/opt/homeassistant/config/custom_components/eldes_alarm/core/eldes_cloud.py", line 118, in renew_token
    response.raise_for_status()
  File "/opt/homeassistant/lib/python3.12/site-packages/aiohttp/client_reqrep.py", line 1161, in raise_for_status
    raise ClientResponseError(
aiohttp.client_exceptions.ClientResponseError: 401, message='', url='https://cloud.eldesalarms.com:8083/api/auth/token'

Probably something wrong with token renewing logic. @augustas2 if you need, I can switch debug mode and collect more logs for it.

[sobakintech]

Eldes was doing some maintenance on their app, so the integration stopped working. After the maintenance ended, this issue started happening to me.

Edit: I think they made some kind of changes to their API that caused this.

[drawgas]

Same here. Issue started today

[simie7]

Who likes to hack - make this change to eldes_alarm/core/eldes_cloud.py:

# replace original function by this hack:
    async def renew_token(self):
        if not self._token_expires_at or datetime.utcnow() < self._token_expires_at:
            _LOGGER.debug("Token is still valid; skipping token refresh.")
            return

        _LOGGER.debug("Token expired, re-authenticating with credentials")

        # Just re-login instead of trying to refresh the token
        await self.login()

# rename original function:
    async def renew_token_old(self):

and restart home assistant. Be careful with each line indentation - Python is sensitive to it.

Problem is, that refresh token API started to fail, maybe some extra header is needed or something else. But login function still works.

It's not fix, but workaround, use at your own risk. I implemented that change ~12h ago, and so far no any ELDES-related error in logs.

[afternet]

Who likes to hack - make this change to eldes_alarm/core/eldes_cloud.py:

# replace original function by this hack:
    async def renew_token(self):
        if not self._token_expires_at or datetime.utcnow() < self._token_expires_at:
            _LOGGER.debug("Token is still valid; skipping token refresh.")
            return

        _LOGGER.debug("Token expired, re-authenticating with credentials")

        # Just re-login instead of trying to refresh the token
        await self.login()

# rename original function:
    async def renew_token_old(self):

and restart home assistant. Be careful with each line indentation - Python is sensitive to it.

Problem is, that refresh token API started to fail, maybe some extra header is needed or something else. But login function still works.

It's not fix, but workaround, use at your own risk. I implemented that change ~12h ago, and so far no any ELDES-related error in logs.

Podrías explicar exactamente que líneas eliminar? Todas? Solo algunas?
Me resulta confusa la modificación.
Un millón de gracias.

Could you explain exactly which lines to delete? All of them? Just some?
I find the modification confusing.
Thanks a million.

[sobakintech]

@simie7 After implementing your workaround, my alarm panel seems to be randomly beeping (as if the entrance door was opened). I'll disable the Eldes integration in Home Assistant for 24 hours and see if it beeps randomly again. It might also just be a coincidence and something is just wrong with my alarm system lol

[triumfas]

Workaround works without issues so far. Thanks!

[afternet]

La solución alternativa funciona sin problemas hasta ahora. ¡Gracias!

SERÍA POSIBLE QUE PUBLIQUEIS UN PANTALLAZO DEL RESULTADO FINAL DEL ARCHIVO?
LA INFO PARECE CONFUSA, DISCULPAD MI TORPEZA.
NO SE SI ES ESTO LO QUE SE REQUIERE, PERO SI NO ME HE EQUIVOCADO, ASÍ NO FUNCIONA.

Would it be possible for you to post a screenshot of the final output of the file?
The information seems confusing, please excuse my clumsiness.
I DON'T KNOW IF THIS IS WHAT IS REQUIRED, BUT IF I'M NOT MISTAKEN, IT DOESN'T WORK LIKE THIS.

[triumfas]

@afternet you need to add renew_token(self) and rename existing one.

    async def renew_token(self):
        if not self._token_expires_at or datetime.utcnow() < self._token_expires_at:
            _LOGGER.debug("Token is still valid; skipping token refresh.")
            return

        _LOGGER.debug("Token expired, re-authenticating with credentials")

        # Just re-login instead of trying to refresh the token
        await self.login()

    async def renew_token_old(self):
        if not self._token_expires_at or datetime.utcnow() < self._token_expires_at:
            _LOGGER.debug("Token is still valid; skipping token refresh.")
            return

        self.headers["Authorization"] = f"Bearer {self._refresh_token}"
        url = f"{API_URL}{API_PATHS['AUTH']}token"

        try:
            async with async_timeout.timeout(self.timeout):
                response = await self._http_session.get(url, headers=self.headers)

            response.raise_for_status()
            result = await response.json()

            _LOGGER.debug("Token successfully refreshed: %s", result)
            return await self._setOAuthHeader(result)

        except aiohttp.ClientResponseError as err:
            _LOGGER.error("Token refresh failed: %s", err)
            raise

        except Exception as e:
            _LOGGER.error("Unexpected error during token refresh: %s", e)
            raise

[afternet]

COPIANDO Y PEGANDO ESTO ME ESTÁ FUNCIONANDO BIEN DURANTE UNA HORA, ANTES ERA IMPOSIBLE.
MUCHAS GRACIAS POR LA ACLARACIÓN.

COPYING AND PASTING THIS HAS BEEN WORKING FINE FOR ME FOR AN HOUR, WHILE IT WAS IMPOSSIBLE BEFORE.
THANK YOU SO MUCH FOR THE CLARIFICATION.

    async def renew_token(self):
        if not self._token_expires_at or datetime.utcnow() < self._token_expires_at:
            _LOGGER.debug("Token is still valid; skipping token refresh.")
            return

        _LOGGER.debug("Token expired, re-authenticating with credentials")

        # Just re-login instead of trying to refresh the token
        await self.login()

    async def renew_token_old(self):
        if not self._token_expires_at or datetime.utcnow() < self._token_expires_at:
            _LOGGER.debug("Token is still valid; skipping token refresh.")
            return

        self.headers["Authorization"] = f"Bearer {self._refresh_token}"
        url = f"{API_URL}{API_PATHS['AUTH']}token"

        try:
            async with async_timeout.timeout(self.timeout):
                response = await self._http_session.get(url, headers=self.headers)

            response.raise_for_status()
            result = await response.json()

            _LOGGER.debug("Token successfully refreshed: %s", result)
            return await self._setOAuthHeader(result)

        except aiohttp.ClientResponseError as err:
            _LOGGER.error("Token refresh failed: %s", err)
            raise

        except Exception as e:
            _LOGGER.error("Unexpected error during token refresh: %s", e)
            raise