You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
After the latest update of the ebusd-configuration (PR john30/ebusd-configuration#316) many Home Assistant users experienced heat pumps, which were shutting off (after 5 to 10 minutes) although the energy integral was still negative. Because of that the configuration changes have been reverted.
During debugging (see john30/ebusd#1205) we found out that ebusdpy and the MQTT integration actively poll read messages and that the new configuration contains read values, which should not be read by a second master (because it will prematurely shut off the heat pump). The theory was successfully tested with the MQTT integration, which is why I am opening this issue.
This part of ebusdpy actively polls read messages:
After the latest update of the ebusd-configuration (PR john30/ebusd-configuration#316) many Home Assistant users experienced heat pumps, which were shutting off (after 5 to 10 minutes) although the energy integral was still negative. Because of that the configuration changes have been reverted.
During debugging (see john30/ebusd#1205) we found out that ebusdpy and the MQTT integration actively poll read messages and that the new configuration contains read values, which should not be read by a second master (because it will prematurely shut off the heat pump). The theory was successfully tested with the MQTT integration, which is why I am opening this issue.
This part of ebusdpy actively polls read messages:
https://github.com/CrazYoshi/ebusdpy/blob/977fe2d795e0521a0a99d33f19f2cf8ba28e18cb/ebusdpy/ebusdpy.py#L30C1-L31C58
Would it be possible to make active polling optional to prevent such system crashes?
The text was updated successfully, but these errors were encountered: