Well, that's a day and a half of my life gone

OK, so I have a b-hyve reticulation controller which, until now, had not been integrated into HA. The issue was forced on me when I switched from mains-water to tank water, which uses a pump to supply. The problem was, how to start the pump when the controller starts a watering program? The controller has a “pump start” output but it is a 24VAC signal - no good for anything i,o,t. I didn’t have a relay at hand and live in a small rural town, so getting one wasn’t going to happen quickly.

Google is my friend and I came across the b-hyve integration for HA. It looked like it would do the job, but it was recommended to use HACS to install it. So, I went through that process, relatively unscathed, and within a couple of hours I had the integration up and working and was able to start my pump as required. Yay!

The real problem started when I was checking some logs as a part of getting the b-hyve integration working. There were so many error messages!


The schema for my Tasmota-loaded lights informed me that “value_template” is deprecated and I have to use “state_value_template” The fix? After a couple of hours of Googling I found a snippet that said that I need to upgrade the Tasmota firmware to a recent version (mine were 8.x.x and there was even a 7.x.x!) To do this ota required a 2-step upgrade, firstly loading a minimal version, then the latest version - for all 18 globes that were affected. Once done I had to setoption 0, to disable auto discovery, install the Tasmota integration and let it find and configure the globes. Yes, this worked and one error in the logs was gone.

Next was a modbus error which basically said that my sensors could not use the “data_type: int”. The problem was, none of my sensors had that line included in their config. More Googling and finally I included the line “data_type: int16” for each sensor and that got rid of another error.

Next was another modbus error which said that the modbus gateway could not be found - even though everything was working perfectly and all sensors were updating! More Googling and the fix was to add the lines:

retry_on_empty: true
retries: 10

and that error went away…

Next came some errors concerning three lines that were in my zigbee2mqtt config and should not have been. Two of the lines deleted with out issue, but one, “new_api: true” (or false, for that matter) would not delete without throwing up a whole bunch of additional errors that would not allow me to even save the changed config file.

Considering that another of my errors concerned zigbee2mqtt not being able to report the link quality of my door locks, I decided to bite the bullet and uninstall/reinstall the add-on. All good. Got rid of the offending line in the config and fixed my link quality issue (after having to re-join all of my zigbee devices, of course…)

That only leaves an ESPHome error which is simply because the devices are currently powered-down and a web hook error, which is the HA app on my phone. What a drama and none of it due to any changes that I had made. All down to changes in HA and/or it’s add-ons over the recent releases. It really shouldn’t be this hard.

1 Like