Modbus timeout since 313.0

[isarrider]

Describe the bug

Since 313.0 I get Modbus timeouts
(Using an ABL eMH1, some commands need to be send multiple times?)

As this did not happen before and I saw some Modbus Timeout related changes,
I thought I report it here...

Trace log of [abl] and [lp-1] attached...

Steps to reproduce

v 313 with a Waveshare USB to RS485 and a eMH1.
Never saw a timeout with the Waveshare USB to RS485 in the 312 version

evcc-20260802-105321-trace.log

Configuration details

{
  "charger": [
    {"id":2,"name":"db:2","type":"template","config":{"id":1,"template":"abl","baudrate":38400,"comset":"8E1","device":"/dev/ttyACM0","modbus":"rs485serial","port":502,"timeout":10000000000},"deviceProduct":"ABL eMH1"}
  ],
  "experimental": true,
  "influx": {"database":"evcc","insecure":false,"org":"Garage","password":"","token":"***","url":"http://172.18.0.2:8428","user":""},
  "interval": 30,
  "loadpoints": [
    {"id":3,"name":"db:3","batteryBoostLimit":100,"charger":"db:2","defaultMode":"minpv","limitEnergy":0,"limitSoc":0,"maxCurrent":32,"minCurrent":6,"minSoc":0,"phasesConfigured":1,"planEnergy":0,"planPrecondition":0,"planStrategy":{"continuous":false,"precondition":0},"planTime":"0001-01-01T00:00:00Z","priority":0,"smartCostLimit":null,"smartFeedInPriorityLimit":null,"soc":{"estimate":true,"poll":{"interval":3600000000000,"mode":"charging"}},"thresholds":{"disable":{"delay":180000000000,"threshold":0},"enable":{"delay":60000000000,"threshold":0}},"title":"Wallbox","ui":{"maxTemp":0,"minTemp":0}}
  ],
  "meter": [
    {"id":1,"name":"db:1","type":"template","config":{"template":"vzlogger","cache":10000000000,"energyscale":"0.001","energyuuid":"b8f36f29-7b6c-493b-9e6f-66fcaca2fe4a","host":"172.18.0.3","port":11111,"returnenergyuuid":"eccbe94d-7f8d-4c49-8ef9-d53b84f25059","scale":"1","usage":"grid","uuid":"***"},"deviceProduct":"vzlogger"}
  ],
  "mqtt": {"broker":"","caCert":"","clientCert":"","clientID":"","clientKey":"","insecure":false,"password":"","topic":"evcc","user":""},
  "residualPower": 0,
  "shm": {"deviceId":"","deviceSerial":"","vendorId":""},
  "site": {"aux":null,"battery":null,"consumer":null,"ext":null,"grid":"db:1","pv":null,"title":"Garage"},
  "tariff": [
    {"id":4,"name":"db:4","type":"template","config":{"template":"fixed","price":0.07780000000000001},"deviceProduct":"Fixed Price"},
    {"id":7,"name":"db:7","type":"template","config":{"template":"fixed","price":0.2462},"deviceProduct":"Fixed Price"},
    {"id":5,"name":"db:5","type":"template","config":{"template":"grünstromindex","token":"***","zip":"***"},"deviceProduct":"Grünstromindex"},
    {"id":6,"name":"db:6","type":"template","config":{"template":"open-meteo","ac":"1000","alphatemp":"-0.004","az":-10,"de":"0","dec":10,"dm":"0","efficiency":"100","interval":"1h","kwp":4.5,"lat":"***","lon":"***","rossmodel":"0.0342"},"deviceProduct":"Open-Meteo","deviceTitle":"PV-Forecast"}
  ],
  "tariffRefs": {"co2":"db:5","feedIn":"db:4","grid":"db:7","planner":"","solar":["db:6"],"temperature":""}
}

Log details

[lp-1  ] DEBUG 2026/08/02 09:29:20 charger status: A
[abl   ] TRACE 2026/08/02 09:29:20 modbus: send 3a 30 31 30 33 30 30 30 46 30 30 30 35 45 38 0d 0a
[abl   ] TRACE 2026/08/02 09:29:20 modbus: recv 3e 30 31 30 33 30 41 30 46 33 46 30 32 31 35 30 32 31 35 30 33 45 38 30 32 31 35 37 34 0d 0a
[abl   ] TRACE 2026/08/02 09:29:20 modbus: send 3a 30 31 30 33 30 30 30 46 30 30 30 35 45 38 0d 0a
[abl   ] TRACE 2026/08/02 09:29:20 modbus: recv 3e 30 31 30 33 30 41 30 46 33 46 30 32 31 35 30 32 31 35 30 33 45 38 30 32 31 35 37 34 0d 0a
[abl   ] TRACE 2026/08/02 09:29:40 modbus: send 3a 30 31 30 33 30 30 32 45 30 30 30 35 43 39 0d 0a
[abl   ] TRACE 2026/08/02 09:29:50 modbus: send 3a 30 31 30 33 30 30 32 45 30 30 30 35 43 39 0d 0a
[abl   ] TRACE 2026/08/02 09:29:50 modbus: recv 3e 30 31 30 33 30 41 32 45 41 31 31 32 31 35 30 33 45 38 30 33 45 38 30 33 45 38 33 42 0d 0a
[lp-1  ] DEBUG 2026/08/02 09:29:50 charge power: 0W
[abl   ] TRACE 2026/08/02 09:29:50 modbus: send 3a 30 31 30 33 30 30 32 45 30 30 30 35 43 39 0d 0a
[abl   ] TRACE 2026/08/02 09:29:50 modbus: recv 3e 30 31 30 33 30 41 32 45 41 31 31 32 31 35 30 33 45 38 30 33 45 38 30 33 45 38 33 42 0d 0a
[abl   ] TRACE 2026/08/02 09:29:50 modbus: send 3a 30 31 30 33 30 30 32 45 30 30 30 35 43 39 0d 0a
[abl   ] TRACE 2026/08/02 09:29:50 modbus: recv 3e 30 31 30 33 30 41 32 45 41 31 31 32 31 35 30 33 45 38 30 33 45 38 30 33 45 38 33 42 0d 0a
[lp-1  ] DEBUG 2026/08/02 09:29:50 charge currents: [0 0 0]A
[abl   ] TRACE 2026/08/02 09:29:50 modbus: send 3a 30 31 30 33 30 30 30 34 30 30 30 31 46 37 0d 0a
[abl   ] TRACE 2026/08/02 09:29:50 modbus: recv 3e 30 31 30 33 30 32 30 34 41 31 35 35 0d 0a
[abl   ] TRACE 2026/08/02 09:29:50 modbus: send 3a 30 31 30 33 30 30 30 34 30 30 30 31 46 37 0d 0a
[abl   ] TRACE 2026/08/02 09:29:50 modbus: recv 3e 30 31 30 33 30 32 30 34 41 31 35 35 0d 0a
[lp-1  ] DEBUG 2026/08/02 09:29:50 charger status: A
[abl   ] TRACE 2026/08/02 09:29:50 modbus: send 3a 30 31 30 33 30 30 30 46 30 30 30 35 45 38 0d 0a
[abl   ] TRACE 2026/08/02 09:29:50 modbus: recv 3e 30 31 30 33 30 41 30 46 33 46 30 32 31 35 30 32 31 35 30 33 45 38 30 32 31 35 37 34 0d 0a
[abl   ] TRACE 2026/08/02 09:29:50 modbus: send 3a 30 31 30 33 30 30 30 46 30 30 30 35 45 38 0d 0a
[abl   ] TRACE 2026/08/02 09:29:50 modbus: recv 3e 30 31 30 33 30 41 30 46 33 46 30 32 31 35 30 32 31 35 30 33 45 38 30 32 31 35 37 34 0d 0a
[abl   ] TRACE 2026/08/02 09:30:10 modbus: send 3a 30 31 30 33 30 30 32 45 30 30 30 35 43 39 0d 0a
[abl   ] TRACE 2026/08/02 09:30:20 modbus: send 3a 30 31 30 33 30 30 32 45 30 30 30 35 43 39 0d 0a
[lp-1  ] ERROR 2026/08/02 09:30:30 charge power: serial: timeout
[abl   ] TRACE 2026/08/02 09:30:30 modbus: send 3a 30 31 30 33 30 30 32 45 30 30 30 35 43 39 0d 0a
[abl   ] TRACE 2026/08/02 09:30:30 modbus: recv 3e 30 31 30 33 30 41 32 45 41 31 31 32 31 35 30 33 45 38 30 33 45 38 30 33 45 38 33 42 0d 0a
[abl   ] TRACE 2026/08/02 09:30:30 modbus: send 3a 30 31 30 33 30 30 32 45 30 30 30 35 43 39 0d 0a
[abl   ] TRACE 2026/08/02 09:30:30 modbus: recv 3e 30 31 30 33 30 41 32 45 41 31 31 32 31 35 30 33 45 38 30 33 45 38 30 33 45 38 33 42 0d 0a
[lp-1  ] DEBUG 2026/08/02 09:30:30 charge currents: [0 0 0]A
[abl   ] TRACE 2026/08/02 09:30:31 modbus: send 3a 30 31 30 33 30 30 30 34 30 30 30 31 46 37 0d 0a
[abl   ] TRACE 2026/08/02 09:30:31 modbus: recv 3e 30 31 30 33 30 32 30 34 41 31 35 35 0d 0a
[abl   ] TRACE 2026/08/02 09:30:31 modbus: send 3a 30 31 30 33 30 30 30 34 30 30 30 31 46 37 0d 0a
[abl   ] TRACE 2026/08/02 09:30:31 modbus: recv 3e 30 31 30 33 30 32 30 34 41 31 35 35 0d 0a
[lp-1  ] DEBUG 2026/08/02 09:30:31 charger status: A
[abl   ] TRACE 2026/08/02 09:30:31 modbus: send 3a 30 31 30 33 30 30 30 46 30 30 30 35 45 38 0d 0a
[abl   ] TRACE 2026/08/02 09:30:31 modbus: recv 3e 30 31 30 33 30 41 30 46 33 46 30 32 31 35 30 32 31 35 30 33 45 38 30 32 31 35 37 34 0d 0a
[abl   ] TRACE 2026/08/02 09:30:31 modbus: send 3a 30 31 30 33 30 30 30 46 30 30 30 35 45 38 0d 0a
[abl   ] TRACE 2026/08/02 09:30:31 modbus: recv 3e 30 31 30 33 30 41 30 46 33 46 30 32 31 35 30 32 31 35 30 33 45 38 30 32 31 35 37 34 0d 0a
[abl   ] TRACE 2026/08/02 09:30:40 modbus: send 3a 30 31 30 33 30 30 32 45 30 30 30 35 43 39 0d 0a
[abl   ] TRACE 2026/08/02 09:30:40 modbus: recv 3e 30 31 30 33 30 41 32 45 41 31 31 32 31 35 30 33 45 38 30 33 45 38 30 33 45 38 33 42 0d 0a
[abl   ] TRACE 2026/08/02 09:30:40 modbus: send 3a 30 31 30 33 30 30 32 45 30 30 30 35 43 39 0d 0a
[abl   ] TRACE 2026/08/02 09:30:40 modbus: recv 3e 30 31 30 33 30 41 32 45 41 31 31 32 31 35 30 33 45 38 30 33 45 38 30 33 45 38 33 42 0d 0a
[lp-1  ] DEBUG 2026/08/02 09:30:40 charge power: 0W
[abl   ] TRACE 2026/08/02 09:30:40 modbus: send 3a 30 31 30 33 30 30 32 45 30 30 30 35 43 39 0d 0a
[abl   ] TRACE 2026/08/02 09:30:40 modbus: recv 3e 30 31 30 33 30 41 32 45 41 31 31 32 31 35 30 33 45 38 30 33 45 38 30 33 45 38 33 42 0d 0a
[abl   ] TRACE 2026/08/02 09:30:40 modbus: send 3a 30 31 30 33 30 30 32 45 30 30 30 35 43 39 0d 0a
[abl   ] TRACE 2026/08/02 09:30:40 modbus: recv 3e 30 31 30 33 30 41 32 45 41 31 31 32 31 35 30 33 45 38 30 33 45 38 30 33 45 38 33 42 0d 0a
[lp-1  ] DEBUG 2026/08/02 09:30:40 charge currents: [0 0 0]A
[abl   ] TRACE 2026/08/02 09:30:40 modbus: send 3a 30 31 30 33 30 30 30 34 30 30 30 31 46 37 0d 0a
[abl   ] TRACE 2026/08/02 09:30:40 modbus: recv 3e 30 31 30 33 30 32 30 34 41 31 35 35 0d 0a
[abl   ] TRACE 2026/08/02 09:30:40 modbus: send 3a 30 31 30 33 30 30 30 34 30 30 30 31 46 37 0d 0a
[abl   ] TRACE 2026/08/02 09:30:40 modbus: recv 3e 30 31 30 33 30 32 30 34 41 31 35 35 0d 0a

What type of operating system or environment does evcc run on?

Docker container

External automation

• I have made sure that no external automation like HomeAssistant or Node-RED is active or accessing any of the mentioned devices when this issue occurs.

Nightly build

• I have verified that the issue is reproducible with the latest nightly build

Version

313.0 and 313.1

[github-actions]

Thanks for the detailed trace log.

Analysis

Looking at the trace, the [abl] connection is configured with timeout: 10000000000 (10s) and abl (charger/abl.go) always performs a double read/write for every register access (get()/set() send the request twice, discarding the first response). When a single response is missed on the RS485 bus, each of the two reads can independently wait out the full 10s timeout, so a single dropped frame can stall for up to ~20s and surface as serial: timeout — which matches the gaps in your log (e.g. 09:30:10 → 09:30:20 → error at 09:30:30).

I checked the Modbus-related changes between 0.312.1 and 0.313.0 and the only one touching the timeout/delay wiring is #31996 (Modbus: unify delay and timeout settings). However, the guard that applies the configured timeout (if timeout > 0 { conn.Timeout(timeout) } in util/modbus/connection.go) and the deprecated timeout parameter for the abl template are functionally unchanged by that refactor, so I can't confirm it as the actual cause with the evidence I have.

That leaves an open question of what's actually causing the RS485 bus to occasionally miss a response since upgrading to 313.

Missing information

To narrow this down, could you provide:

• Was the timeout: 10s value on the abl device something you set deliberately, or has it been in your config for a long time? Does lowering it (or removing the deprecated timeout param entirely and relying on the default) change anything?
• Any USB/serial errors around the timeout timestamps in dmesg/journalctl on the host — this would point at the /dev/ttyACM0 adapter/driver rather than evcc itself?
• Did anything else change on the host besides the evcc version (kernel/driver update, different USB port/cable, container runtime)?

Without evidence that the same 10s-timeout config actually behaves differently between 312 and 313 (rather than an intermittent RS485/USB issue), I can't confirm a software regression yet.

🤖 Generated with Claude Code

[isarrider]

Timeout was default, is untouched...

dmesg throws no errors, so the adapter seems fine...

Nothing else changed except a full system update,
but that was done already a couple of days before...
(Kernel Linux Pi 6.18.39+rpt-rpi-v8 and Docker version 29.7.1, build e9452d6 now)

[isarrider]

Ok, removing the timeout seems to help at least, no timeout since 11:30.
I will report later toda back...

[isarrider]

Still no timeout - seems removing the timeout really fixed it...