Frequent lost connection to Huawei Inverters

[Taraman17]

Hi,

I am running the custom Huawei Solar integration on HA.
I have 2 cascaded inverters, that are addressed via Modbus over LAN. Each has it's own slave ID.

The Integration regularly loses the connection to the inverters. Or more specific, it seems that single messages are lost.

Now I discovered a log entry, that points towards pymodbus:

2026-04-07 11:11:06.999 ERROR (MainThread) [pymodbus.logging] ERROR: request ask for transaction_id=13810 but got id=13808, Skipping.
>>>>> send: 0x35 0xea 0x0 0x0 0x0 0x6 0x2 0x3 0x7d 0x0 0x0 0x14
>>>>> recv: 0x35 0xea 0x0 0x0 0x0 0x2b 0x2 0x3 0x28 0x0 0x6 0x0 0x2 0x0 0x7 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x2f 0x61 0x6 0xf2 0x4 0x62 0x0 0x0 0x0 0x0 extra data: 
>>>>> send: 0x35 0xeb 0x0 0x0 0x0 0x6 0x2 0x3 0x7d 0x40 0x0 0x34
>>>>> recv: 0x35 0xeb 0x0 0x0 0x0 0x6b 0x2 0x3 0x68 0x0 0x0 0x7 0xa9 0xf 0xcc 0xf 0xd0 0xf 0xce 0x9 0x21 0x9 0x19 0x9 0x26 0x0 0x0 0xa 0x77 0x0 0x0 0xa 0x83 0x0 0x0 0xa 0x98 0x0 0x0 0x7 0x6f 0x0 0x0 0x7 0x62 0xff 0xff 0xff 0xfc 0x3 0xe8 0x13 0x8a 0x25 0xa6 0x1 0xc6 0xb 0xb8 0x2 0x0 0x0 0x0 0x69 0xd4 0xac 0x93 0x69 0xd4 0xd 0x4f 0x0 0x0 0x7 0x64 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0xff 0xff 0xff 0xf8 0x0 0x0 0x0 0x0 0x0 0x9 0xea 0xdf 0x0 0xa 0xce 0xfe 0x69 0xd4 0xe6 0x13 0x0 0x0 0x0 0x20 0x0 0x0 0x2 0x0 extra data: 
>>>>> send: 0x35 0xec 0x0 0x0 0x0 0x6 0x1 0x3 0x7d 0x0 0x0 0x14
>>>>> recv: 0x35 0xec 0x0 0x0 0x0 0x2b 0x1 0x3 0x28 0x0 0xa 0x0 0x2 0x0 0x7 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0xea 0x30 0x10 0xc8 0x2 0xfb 0xa 0x74 0x0 0x88 extra data: 
>>>>> send: 0x35 0xed 0x0 0x0 0x0 0x6 0x1 0x3 0x7d 0x40 0x0 0x34
>>>>> recv: 0x35 0xed 0x0 0x0 0x0 0x6b 0x1 0x3 0x68 0x0 0x0 0xe 0x4 0xf 0xc5 0xf 0xc5 0xf 0xcc 0x9 0x21 0x9 0x11 0x9 0x22 0x0 0x0 0x5 0xe3 0x0 0x0 0x5 0xdb 0x0 0x0 0x5 0xf5 0x0 0x0 0x15 0x39 0x0 0x0 0x4 0x19 0xff 0xff 0xff 0xfe 0x3 0xe8 0x13 0x8b 0x27 0x10 0x1 0xef 0xb 0xb8 0x2 0x1 0x0 0x0 0x69 0xd4 0xb6 0x69 0xff 0xff 0xff 0xff 0x0 0x0 0x4 0x19 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0xff 0xff 0xff 0xfd 0x0 0x2 0x3b 0x6a 0x0 0x1e 0x80 0x59 0x0 0x20 0xf0 0xb9 0x69 0xd4 0xe6 0x1d 0x0 0x0 0x0 0x1b 0x0 0x0 0x1 0xba extra data: 
>>>>> send: 0x35 0xee 0x0 0x0 0x0 0x6 0x1 0x3 0x90 0xed 0x0 0x25
>>>>> recv: 0x35 0xee 0x0 0x0 0x0 0x4d 0x1 0x3 0x4a 0x0 0x0 0x9 0x22 0x0 0x0 0x9 0x27 0x0 0x0 0x9 0x2c 0x0 0x0 0x1 0x59 0x0 0x0 0x1 0x25 0x0 0x0 0x1 0x59 0x0 0x0 0x8 0x7d 0x0 0x0 0x1 0xe8 0xfc 0x55 0x13 0x8a 0x0 0x11 0x9f 0x7d 0x0 0x18 0x2d 0x91 0x0 0x0 0x0 0x0 0x0 0x1 0x0 0x0 0xf 0xd7 0x0 0x0 0xf 0xde 0x0 0x0 0xf 0xda 0x0 0x0 0x2 0xf8 0x0 0x0 0x2 0x6b 0x0 0x0 0x3 0x1a extra data: 
>>>>> send: 0x35 0xef 0x0 0x0 0x0 0x6 0x1 0x3 0x93 0x80 0x0 0x1c
>>>>> recv: 0x35 0xef 0x0 0x0 0x0 0x3b 0x1 0x3 0x38 0x1 0xa4 0x0 0x0 0x0 0x2 0x1d 0xf8 0x0 0x21 0x0 0x0 0x9 0xf2 0x0 0x0 0x13 0x88 0x0 0x0 0x13 0x88 0x0 0x0 0x13 0x88 0x0 0x0 0x13 0x88 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x7 0x9d 0x84 0x0 0x7 0x8c 0xe2 0x0 0x0 0x1 0x27 0x0 0x0 0x0 0x6a extra data: 
>>>>> send: 0x35 0xf0 0x0 0x0 0x0 0x6 0x2 0x3 0x7d 0x0 0x0 0x14
>>>>> Repeating....
>>>>> recv: 0x35 0xf0 0x0 0x0 0x0 0x3 0x2 0x83 0x6 extra data: 
>>>>> send: 0x35 0xf1 0x0 0x0 0x0 0x6 0x2 0x3 0x7d 0x0 0x0 0x14
>>>>> recv: 0x35 0xf1 0x0 0x0 0x0 0x3 0x2 0x83 0x6 extra data: 
>>>>> send: 0x35 0xf2 0x0 0x0 0x0 0x6 0x2 0x3 0x7d 0x0 0x0 0x14
>>>>> recv: 0x35 0xf2 0x0 0x0 0x0 0x3 0x2 0x83 0x6 extra data: 
>>>>> recv: 0x35 0xf0 0x0 0x0 0x0 0x2b 0x2 0x3 0x28 0x0 0x6 0x0 0x2 0x0 0x7 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x2f 0x61 0x6 0xf1 0x4 0x66 0x0 0x0 0x0 0x0 extra data: 

So the question here is, if this is worth a bug report here or elsewhere.

Best Regards
Markus

[janiversen]

It is hard to tell, why you get this error, because you have not included any log lines before the error.

You have also not written which version of pymodbus you are using, I remember there was a bug fix related to transaction ids in the newest release.

The error is because your app send a request but received an old response.

[janiversen]

I run a Sun2000 with the standard modbus without problems.

[janiversen]

Just analyzed the data leading up to the problem

send: 0x35 0xf0 0x0 0x0 0x0 0x6 0x2 0x3 0x7d 0x0 0x0 0x14
Repeating....

The send is repeated because there are no response, within the timeout you have defined.

recv: 0x35 0xf0 0x0 0x0 0x0 0x3 0x2 0x83 0x6 extra data:

Then you get a response, and the app continues sending requests with a higher tid.

send: 0x35 0xf1 0x0 0x0 0x0 0x6 0x2 0x3 0x7d 0x0 0x0 0x14
recv: 0x35 0xf1 0x0 0x0 0x0 0x3 0x2 0x83 0x6 extra data:
send: 0x35 0xf2 0x0 0x0 0x0 0x6 0x2 0x3 0x7d 0x0 0x0 0x14
recv: 0x35 0xf2 0x0 0x0 0x0 0x3 0x2 0x83 0x6 extra data:
recv: 0x35 0xf0 0x0 0x0 0x0 0x2b 0x2 0x3 0x28 0x0 0x6 0x0 0x2 0x0 0x7 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x0 0x2f 0x61 0x6 0xf1 0x4 0x66 0x0 0x0 0x0 0x0 extra data:

Here the device returns with an old tid, probably due to the retry earlier.

So pymodbus correctly detects a wrong tid. I think the current version, makes a log e try but continues.

[janiversen]

If you want to ignore it, you need to catch the exception and handle it.

[Taraman17]

Thank you for the analysis.
So that means that raising the timeout might help and else the integration needs to handle the missing response better?

[janiversen]

Correct, in all cases I always try to use a try/except around library calls to make my own handling, doing that allows you to e.g. ignore it and continue.