Craig Dean

Yeah, we tried resetting the new mode flag followed by a comms board reboot; and I previously tried a cold boot whilst in the new mode. I've suggested the next...

OK, I'll hold on to try that first. I don't want it to be our issue as I've got a really good line to their engineers right now, and they're...

OK, I ran using that and I didn't see any `ModbusIOException` in the output. ``` 2024-01-23 18:09:33.247 DEBUG (MainThread) [pymodbus.logging] Connecting to 192.168.89.177:1502. 2024-01-23 18:09:33.247 DEBUG (MainThread) [pymodbus.logging] Connecting comm...

Yeah, it doesn't also log it. I appreciate that is the HA Devs choice. Is it considered anti-pattern to also put the error in the log? Personally, I'd put the...

Yeah I’m just starting to read up on the Sun Spec. I’ll see if I can get more info, but I’m not hugely optimistic. Ofc I’ll share it if I...

The plot thickens, I sent raw data to the port and could see that it responded with zero-length data. I changed the unit identifier to '02' and voila, I got...

OK, I just deleted the hub and added a new one with Device ID 2. The setup still fails, but I'm getting a lot more logs now: ```log 2024-01-24 09:40:19.205...

Great news! It turns out that they set the Device ID for _both_ the master and slave inverters to '2'!!! I've queried why they did that - I hope it...

> Scanning all possible modbus addresses isn't really practical. With minimum timeouts would take at least 12 minutes, and nobody is going to wait that long for HA to finish...

> There are commercial-size installations using this integration, so we can't make any assumptions about small sites. Agreed, but I'm not. As stated above, the 32-device limit is the same...