Hello,
I’m trying the external integration ebusd by @Will_71 (ghcr.io/william-de71/gladys-ebusd:0.1.0) and the test fails even though everything seems correct on my end. Can you tell me if the problem comes from the integration?
Environment
- Gladys Assistant 5.3.2 on Gladys OS (Raspberry Pi 5, Raspberry Pi OS Trixie 64-bit, flashed with Raspberry Pi Imager). Containers:
gladys,watchtower,ebusd,gladys-ext-william-de71-gladys-ebusd. - ebusd
john30/ebusd26.1.26.1 in Docker on the same Pi (192.168.1.13):
docker run -d --name=ebusd --restart=unless-stopped -p 8888:8888 john30/ebusd --scanconfig -d ens:192.168.1.14:9999 - Adapter: eBUS Adapter Stick C6 (ESP32-C6), firmware 20261109, on Wi-Fi (192.168.1.14), enhanced protocol over TCP. Adapter Status page: “ebusd connected: yes”, “eBUS signal: no signal” (normal, see below).
Adapter settings (Enhanced protocol, TCP):
- Boiler: Vaillant, not yet connected (I’m preparing the installation before connecting it). The eBUS signal is therefore “no signal” for now.
Integration Configuration
Ebusd address: 192.168.1.13 (without http://), port 8888, empty username and secret, “Allow modifications” unchecked. The configuration saves without any issues.
What I Observe
The test displays ebusd does not respond to 192.168.1.13:8888 (element not found) and the badge remains on “Disconnected”:
A few minutes earlier, with the same configuration, another message appeared: “ebusd responds but hears nothing on the eBUS: check the adapter and its wiring”. No errors in docker logs gladys-ext-william-de71-gladys-ebusd (only [gladys-sdk] connected to Gladys).
What Shows That ebusd Is Reachable and Working
docker exec ebusd ebusctl info:
version: ebusd 26.1.26.1
device: 192.168.1.14:9999, TCP, enhanced, firmware 1.1[6b09].1[6b09]
signal: no signal
reconnects: 0
masters: 1
messages: 11
address 31: master #8, ebusd
address 36: slave #8, ebusd
docker exec ebusd ebusctl find:
Broadcast Datetime = no data stored
Broadcast Error = no data stored
Broadcast Id = no data stored
Broadcast IdAnswer = no data stored
Broadcast Signoflife = no data stored
Memory Eeprom = no data stored
Memory Ram = no data stored
Ebusd responds correctly and only knows its 7 internal messages, with no Vaillant devices, since the boiler is not connected.
nc -vz 192.168.1.13 8888from the Pi: succeeded.- From inside the integration container (
net.connectunder Node):172.30.0.1:8888OK and192.168.1.13:8888OK. - ebusd has been running for several hours without reconnection.
My Hypothesis
element not found seems to be a response from ebusd to a command targeting a missing element. I reproduced it manually:
docker exec ebusd ebusctl find -c bai
ERR: element not found
(bai is the Vaillant boiler circuit, which is absent in my case since it is not connected.) However, a simple ebusctl find works and does not return this error. The integration may treat this response as a failure (“ebusd is not responding”) when in fact ebusd is responding and simply indicating that the element does not exist yet.
My questions
- Is it this response (
ERR: element not found, for example when searching for a circuit likebai) that the integration receives? In this case, can it be treated as “no device found” rather than a failure when the bus is empty? - Does the boiler need to be connected and detected by ebusd for the test to pass?
- Is there a way to get more detailed logs in the integration container?
I can provide additional logs or screenshots. Thanks in advance!
it’s working now.