FW 0.9.48 (série E) control_function switch KO

[Clust3r]

Hi i just upgrade from 0.9.43rc to 0.9.48 live the separated topics but I loose my control under HA for DHW boost etc.

When i try to active the control_function switch. Nothing change.

[Arnold-n]

Thanks for notifying. One of the configuration messages is truncated, which may explain your control loss. P1P2MQTT-bridge v0.9.49 for the ESP increases the maximum configuration message size to 1500 and should solve this.

I am somewhat surprised about the longer msg size on your side, if you don't mind, could you (after upgrade) share the (longest) homeassistant/# message(s)? Note: If you changed any of the predefined names to personal ones, these may be visible in the config messages.

[Clust3r]

Yes ATmega allready updated:

[ESP] 2024-05-19 17:16:07 Subscribed to P1P2/W result 5 [ESP] 2024-05-19 17:16:07 Subscribed to P1P2/W/P1P2MQTT/bridge0 result 6 [ESP] 2024-05-19 17:16:07 Subscribed to homeassistant/status [ESP] 2024-05-19 17:16:07 Subscribed to P1P2/P/meter/U/9/Electricity_Power [ESP] 2024-05-19 17:16:07 Subscribed to P1P2/P/meter/U/9/Electricity_Total [ESP] 2024-05-19 17:16:07 Subscribed to P1P2/P/meter/U/9/Electricity_BUH_Power [ESP] 2024-05-19 17:16:07 Subscribed to P1P2/P/meter/U/9/Electricity_BUH_Total [ESP] 2024-05-19 17:16:07 < ESP reboot reason: D0 [ESP] 2024-05-19 17:16:07 < No ethernet hardware detected [MON] 2024-05-19 17:16:07 P1P2Monitor v0.9.46 compiled May 18 2024 15:53:07 bran d 1 modelnr 3 +control NEWP1P2LIB E-series [ESP] 2024-05-19 17:16:07 < Connected to WiFi [MON] 2024-05-19 17:16:07 Reset cause: MCUSR=5 (brown-out) (power-on-reset) [ESP] 2024-05-19 17:16:07 < MQTT client connected [MON] 2024-05-19 17:16:07 P1P2MQTT bridge ATmegaHwID v1.2 [MON] 2024-05-19 17:16:07 Control_id=0xF0 [ESP] 2024-05-19 17:16:07 < Telnet 10.0.10.191 connected [MON] 2024-05-19 17:16:07 Counter_requests=1

Any advise to log all MQTT traffic ? I'm not very familiar with MQTT loging mecanics

[Clust3r]

... I'm sorry just find the "solution" i have made intervention on my LWT circuit 2 days ago and the system was still in "Installer mode" (Code 5678) go back to "Advanced User Mode" Code 1234

got back all the switches

Sorry for the time lost on this.
Regards

[Arnold-n]

No problem at all - this is crucial information for others. My system just continues to operate in installer mode (and drops out of installer mode after an hour). It is interesting to learn that installer mode blocks regular control from a second controller in your system. I'd be interested to know how we can see in the hex data logs that the system is in installer mode, so we can add feedback in HA - this may be useful also for the quiet level discussion.

[Clust3r]

If i can help no problem. I can switch User Mode and monitoring via telnet. ATM tried to but when i change the user mode from User to installer, i didn't found any change in the log at the same time.

Maybe i need to use/dump something specific ?

Regards

[Arnold-n]

Perhaps there is information in packet 0x30 in the raw hex data, which is not decoded into MQTT entities currently. If you could log P1P2/R/# data from telnet (J012) when making the user/installer change or back, that would be helpful.