Inverter 0 unable to read REST data

[dafunkmaster]

Hi All,

Everything seemingly has been working but perhaps not, as I went into the garage today and noticed the battery was on one green light, so not charged. So made me think that HA didn't schedule the dump last night and charge overnight; oh was I right.

This was all working, I've not looked how far back this started happening, but have updated HA and everything to the latest versions.

Add-on: Predbat
Home Battery Prediction and Control

Add-on version: 1.3.2
You are running the latest version of this add-on.
System: Home Assistant OS 16.3 (amd64 / qemux86-64)
Home Assistant Core: 2025.11.1
Home Assistant Supervisor: 2025.11.1

My setup is the GivEnergy 9.5kw + Inverter + Panels using Octopus and have a Zappi car charger. This setup was always working, all the usual things ...(dumps before 23:30, charges when the car is on a IOG rate during the day, etc, etc etc.)

I seem to have found this in my logs:

**** Starting Standalone Predbat ****
2025-11-09 09:29:57.394874: Loading apps.yaml
2025-11-09 09:29:57.433339: Info: Add-on slug is 6adb4f0d_predbat
2025-11-09 09:29:57.440315: Creating task: <coroutine object HAInterface.start at 0x7ff78e0a7e00>
2025-11-09 09:29:57.440757: Info: Starting HA interface
2025-11-09 09:29:57.440783: Info: Start socket for url http://192.168.1.70:8123/api/websocket
2025-11-09 09:29:57.443018: Info: Web Socket active
2025-11-09 09:29:58.440712: Creating task: <coroutine object HAHistory.start at 0x7ff78e2e8630>
2025-11-09 09:29:58.441021: Info: Starting HAHistory
2025-11-09 09:29:58.441100: Info: HAHistory: Pruning history data
2025-11-09 09:29:59.441659: Creating task: <coroutine object WebInterface.start at 0x7ff78e35b450>
Web interface started
2025-11-09 09:30:00.725775: Creating task: <coroutine object SolarAPI.start at 0x7ff78e297c40>
Watching ['./components.py', './inverter.py', './hass.py', './gecloud.py', './fox.py', './unit_test.py', './octopus.py', './utils.py', './plugin_system.py', './prediction.py', './db_engine.py', './config.py', './execute.py', './output.py', './apps.yaml', './web_mcp.py', './predheat.py', './alertfeed.py', './energydataservice.py', './solcast.py', './futurerate.py', './ohme.py', './userinterface.py', './ha.py', './web.py', './fetch.py', './download.py', './predbat.py', './compare.py', './db_manager.py', './plan.py', './carbon.py', './web_helper.py'] for changes
2025-11-09 09:30:03.402024: Warn: Historical day 7 has 570 minutes of gap in the data, filled from 0.69 kWh to make new average 1.15 kWh (percent 60%)
2025-11-09 09:30:03.424470: Error: Exception raised HTTPConnectionPool(host='192.168.1.70', port=6345): Max retries exceeded with url: /readData (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7ff78e1b64b0>: Failed to establish a new connection: [Errno 111] Connection refused'))
2025-11-09 09:30:03.424498: Warn: Inverter 0 unable to read REST data from http://192.168.1.70:6345/readData - REST will be disabled
2025-11-09 09:30:03.427544: Warn: record_status Inverter 0 unable to read REST data from http://192.168.1.70:6345/readData - REST will be disabled
2025-11-09 09:30:03.427584: Warn: Inverter control auto restart trigger: REST read failure command []
2025-11-09 09:30:03.427594: Info: auto_restart not defined in apps.yaml, Predbat can't auto-restart inverter control
2025-11-09 09:30:03.437571: Error: Exception raised HTTPConnectionPool(host='192.168.1.70', port=6345): Max retries exceeded with url: /readData (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7ff78e1b5d60>: Failed to establish a new connection: [Errno 111] Connection refused'))

---

When I navigate to: http://192.168.1.70:8123/api/websocket

I get this:
o WebSocket UPGRADE hdr: None
Can "Upgrade" only to "WebSocket".

When I nagivate to http://192.168.1.70:6345/readData

I get all the battery data ...

I am not sure where to go look

[gcoan]

Try a full restart of Home Assistant and all the addons

[dafunkmaster]

Yeah tried that, actually rebooted the whole docker image that's running HA

[gcoan]

the failure is from predbat not being able to make a REST call to givtcp

I would check what is in the givtcp addon log, that might indicate the problem.

But a quicker fix could be to upgrade to the latest givtcp 3.5 that was released today, I've been using dev versions of it for a number of weeks and it's a lot more reliable on REST calls.

Another fix is to comment out the givtcp_rest lines from apps.yaml and use the HA givtcp entities.

I have had similar issues with REST and givtcp, for months I wasn't able to use REST and had to use the HA entities. Latest Givtcp has for me resolved these issues

[dafunkmaster]

Let me update, saw it available this morning and will let you know