Modbus connection errors: how to manage from the high level

Configuration
1

Hi guys,
I’m observing frequent modbus communication errors and I’d like to understand how to manage them from the automation level.
Such an error is becoming really frequent, so a good point could be understanding why it happens, but, of course, also how to manage and restart the system to make the connection available again.
This is the error:

023-12-15 16:12:13.898 DEBUG (SyncWorker_4) [homeassistant.components.modbus.modbus] Pymodbus: hub_termica: Error: device: 25 address: 28317 -> Modbus Error: [Connection] Failed to connect[ModbusSerialClient(<pymodbus.framer.rtu_framer.ModbusRtuFramer object at 0x7fbf9a8a6090> baud[9600])]

ModBus_Error

Moreover, I’m registering also a different kind of error, minor I suppose that it is this one:

2023-12-17 15:04:54.708 ERROR (SyncWorker_10) [homeassistant.components.modbus.modbus] Pymodbus: hub_termica: Error: device: 23 address: 46801 -> pymodbus returned isError True 2023-12-17 15:05:27.789 ERROR (SyncWorker_0) [homeassistant.components.modbus.modbus] Pymodbus: hub_termica: Error: device: 20 address: 28311 -> Modbus Error: [Input/Output] No Response received from the remote slave/Unable to decode response

any ideas?

2

I’m assuming that perhaps you are either too fast at polling or you are getting occasional corruption.

I think the modbus options for the modbus component can be modified to slow down or handle errors more gracefully.

I don’t remember the options at the moment. Are you receiving the expected register information? If you are and the error is occasionally occurring, then I suspect one or the other is happening.

3

Thanks for the interest swill008!
With regards the first problem, I understand that it is a USB connection error… I have to understand how to restart the driver in that case.

With regards the second one, I have those random errors on the communication related to different devices (9 devices connected to the hub). This happens after a while.
You say polling time… Maybe.
But I don’t understand how the HA modbus component manages the queue and the channel, so I’m not able to apply a correction.
Is the channel open/closed or is it always open? There is a way to change that behaviour?
My entities requests the state every 15 seconds, is it too fast?

1 Like
4

Hello,

I have same problem you described:
2024-02-10 21:26:09.034 ERROR (SyncWorker_8) [homeassistant.components.modbus.modbus] Pymodbus: SungrowSHx: Error: device: 1 address: 5725 → pymodbus returned isError True

Every 10 seconds, for different addresses.

Have you found a solution for this in the meantime?

Thanks in advance for any feedback.

Home Assistant

Core 2024.2.1
Supervisor 2024.01.1
Operating System 11.5
Frontend 20240207.1
5

I think I answered it myself.
My integration file contains queries for my inverter, which presumably cannot exist there yet because the system is too new:
Annual kWH export, months March to December, etc.