RAS-3M18G3AVG-E Inverter / Restart Integration

[tux1980]

Hi,

I do have a weird issue with this integration. At least once a day I have to restart the Integration because it claims "invalid credentials". So this indicates an authentication error I think:).

Is disturbing and annoying a bit since I'm using automation to control and regulate my ac devices. As soon I restart the integration it's been working again.

Excerpt from the log book:
2024-08-31 12:31:11.957 ERROR (MainThread) [toshiba_ac.device] State reload failed: 2024-08-31 12:31:32.760 ERROR (MainThread) [toshiba_ac.device] State reload failed: 2024-08-31 13:01:11.530 ERROR (MainThread) [toshiba_ac.device] State reload failed: 2024-08-31 13:34:50.603 ERROR (MainThread) [toshiba_ac.device] State reload failed: File "/config/custom_components/toshiba_ac/climate.py", line 197, in async_set_hvac_mode await self._device.set_ac_status(ToshibaAcStatus.ON) File "/usr/local/lib/python3.12/site-packages/toshiba_ac/device/__init__.py", line 269, in set_ac_status File "/usr/local/lib/python3.12/site-packages/toshiba_ac/device/__init__.py", line 259, in send_state_to_ac File "/usr/local/lib/python3.12/site-packages/toshiba_ac/utils/amqp_api.py", line 100, in send_message 2024-08-31 16:01:28.193 ERROR (MainThread) [toshiba_ac.device] State reload failed: 2024-08-31 16:31:37.953 ERROR (MainThread) [toshiba_ac.device] State reload failed: File "/config/custom_components/toshiba_ac/climate.py", line 197, in async_set_hvac_mode await self._device.set_ac_status(ToshibaAcStatus.ON) File "/usr/local/lib/python3.12/site-packages/toshiba_ac/device/__init__.py", line 269, in set_ac_status File "/usr/local/lib/python3.12/site-packages/toshiba_ac/device/__init__.py", line 259, in send_state_to_ac File "/usr/local/lib/python3.12/site-packages/toshiba_ac/utils/amqp_api.py", line 100, in send_message 2024-08-31 16:44:34.929 ERROR (MainThread) [custom_components.toshiba_ac] Unload Toshiba integration 2024-08-31 16:44:35.377 WARNING (MainThread) [custom_components.toshiba_ac] Initial connection failed, trying to get new sas_token... 2024-08-31 17:01:17.277 ERROR (MainThread) [toshiba_ac.device] State reload failed: 2024-08-31 17:31:02.835 ERROR (MainThread) [toshiba_ac.device_manager] Fetching energy consumption failed:

Every hint is very welocme
thanks!
tux1980

[tux1980]

The Problem is obviously that the Token is running out and it will be generate a new Token after restarting the Integration:
"username": "tuxxxxxxxx", "password": "xxxxxxxxxxxxxxx",
"device_id": "5c6afa0d301309d5",
"sas_token": "SharedAccessSignature sr=toshibasmaciothubprod.azure-devices.net%2Fdevices%2Ftux19....."

Any Idea how I can assign a task to ha to generate that token eg all 30 minutes and put in "core.config_entries" file?

thanks

[KaSroka]

What is the expiry date of your sas token? You can check it with cat /config/.storage/core.config_entries | jq '.data.entries[] | select(.domain == "toshiba_ac") | .data.sas_token | split("&") | map(select(startswith("se="))) | .[0] | ltrimstr("se=") | tonumber | todate' in my case I have tokens valid for 100 years.

Also, duplicate of #197

[h4de5]

same here.. my ends in 2124

[tux1980]

cat /config/.storage/core.config_entries | jq '.data.entries[] | select(.domain == "toshiba_ac") | .data.sas_token  | split("&") | map(select(startswith("se="))) | .[0] | ltrimstr("se=") | tonumber | todate'
"2124-08-13T23:42:44Z"

[tux1980]

Yes, same here, but nevertheless. It dows not work - I still have to restart the integration in regular intervals.

Logger: homeassistant.helpers.entity_platform
Quelle: helpers/entity_platform.py:415
Erstmals aufgetreten: 03:00:42 (1 Vorkommnisse)
Zuletzt protokolliert: 03:00:42

toshiba_ac raises exception ConfigEntryNotReady in forwarded platform climate; Instead raise ConfigEntryNotReady before calling async_forward_entry_setups

Dieser Fehler wurde von einer benutzerdefinierten Integration verursacht

Logger: custom_components.toshiba_ac.climate
Quelle: custom_components/toshiba_ac/climate.py:57
Integration: Toshiba AC (Dokumentation, Probleme)
Erstmals aufgetreten: 03:00:42 (1 Vorkommnisse)
Zuletzt protokolliert: 03:00:42

Error during connection to Toshiba server

[tux1980]

This occurred at 3am - now I restarted the Integration again and everything works fine again.

[h4de5]

does something happen at 3am that would make integration to restart? also the logs from above looks like it happens more or less every 30 minutes. in the current version there should be some kind of randomness to prevent DoS like behavior.

[KaSroka]

Can you please enable debug logs and post the log output when issue happens?

To enable logs, please add this to your configuration.yaml

logger:
  default: warning
  logs:
    # log level toshiba-ac
    toshiba_ac: debug
    custom_components.toshiba_ac: debug

[tux1980]

Hi,

I activated the debugging Modus and the outcome we can find here:

home-assistant_toshiba_ac_debug.log

[h4de5]

do you have some kind of automation that reloads the integration regulary?

[tux1980]

Yes, but that did not help as a workaround and I deactivated it yesterday.

[h4de5]

in the logs its this on, right?

[tux1980]

Yes, it is on in the logs. It is weird nevertheless that a manual restart helps to activate the entities again but via an automation the integration didn't start up at all.....

[KaSroka]

I have created #203 that may fix the issue.

[tux1980]

Thanks for merging the Repos. I did the update today and will stay tuned - activated debugging too in the case anything goes south again.

That's the great thing with Open Source....:). If I would have the capabilities as developer, as you, I would contribute myself for sure.
So thanks again both of you!

[tux1980]

Hi. Just testet it and it seemed to working at least the entities are all there now - but now I get the message when e.g. starting cooling modus or something else:
The Action climate/set_hvac_mode could not be executed. Cannot execute method - Pipeline is not running

Same after restart of the Integration.

And the debug Log:
home-assistant_toshiba_ac_2024-09-11T11-41-35.108Z.log

Any Ideas are welcome

[tux1980]

Hello. I just flushed the pipeline yesterday and it worked.

But, unfortunately, the original error with the sas token stays with the new version. After couple of hours the entities are not available any longer and I have to restart my toshiba integration to make it work again.

[tux1980]

Hi, error stays.... :(. I will attach the debug protocol again. Any Idea?
home-assistant_toshiba_ac_2024-09-16T18-23-36.654Z.log

[KaSroka]

Other user reported that it is working with the fix. I can see that there is something wrong with your setup as it shouldn't take that long to setup the integration:

2024-09-15 15:00:14.094 WARNING (MainThread) [homeassistant.components.climate] Setup of climate platform toshiba_ac is taking over 10 seconds.
2024-09-15 15:00:14.094 WARNING (MainThread) [homeassistant.components.select] Setup of select platform toshiba_ac is taking over 10 seconds.
2024-09-15 15:00:14.095 WARNING (MainThread) [homeassistant.components.sensor] Setup of sensor platform toshiba_ac is taking over 10 seconds.
2024-09-15 15:00:14.095 WARNING (MainThread) [homeassistant.components.switch] Setup of switch platform toshiba_ac is taking over 10 seconds.
2024-09-15 15:01:03.789 ERROR (MainThread) [homeassistant.components.climate] Setup of platform toshiba_ac is taking longer than 60 seconds. Startup will proceed without waiting any longer.
2024-09-15 15:01:03.792 ERROR (MainThread) [homeassistant.components.select] Setup of platform toshiba_ac is taking longer than 60 seconds. Startup will proceed without waiting any longer.
2024-09-15 15:01:03.792 ERROR (MainThread) [homeassistant.components.sensor] Setup of platform toshiba_ac is taking longer than 60 seconds. Startup will proceed without waiting any longer.
2024-09-15 15:01:03.793 ERROR (MainThread) [homeassistant.components.switch] Setup of platform toshiba_ac is taking longer than 60 seconds. Startup will proceed without waiting any longer.
2024-09-15 15:01:36.514 ERROR (MainThread) [custom_components.toshiba_ac.climate] Error during connection to Toshiba server

The unloading you are doing is definitely hurting your setup, and I can see other integrations are also failing. Is your internet connection stable, many warnings are suggesting you have problem with connectivity?

Also, there is nothing regarding the SAS token in the attached log.