The issue got fixed and deployed. So will remove the addition platform: linux/armv7
again.
The more you write the more I believe I have a bug in my wmbusmeters volume definition and the correspoding samba share.
Regarding your question of āmissing nanoā to do the configuration after going the docker exec -it wmbusmeters /bin/sh
-route: Thereās also the alternative of accessing the volumes directly. Use sudo su
and then cd /var/lib/docker/volumes/
. You should then see your wmbusmeters-directory, and do the config-setting via nano
directly there.
Btw: Iām running a system not straightly set up via my instruction above (so I also played around a bit), but I do find three wmbusmeters-folders (and only one linked via Samba)ā¦
ā¦so I believe your problem is just the result of a bug in my volume definitions, leading to multiple wmbusmeters volumes, but just one being the correct one. Iāll fix this as well, but might take some time until I find the time.
Samba is still somehow a mystery to me. Sometimes it works, sometimes it doesnāt. Maybe you find out what needs to be changed in the docker-compose.yml, so that I can update the guideline above.
I think we are really close
''wmbusmeters_data" volume is missing within my configuration
at the same time āopenhab_data_wmbusmeters_dataā volume looks to be empty while treated with ālsā commandā¦
Try to find out which is the ārealā volume thatās currently being used by wmbusmeters (and ignore the others which are caused by my bug ), do your configuration there, and then it should work. The fact that the stick is already working is majority of the work.
I know only basic linux
I found this command and it works:
sudo chmod -R 777 /var/lib/docker/volumes/
Hope you sleep well - there is no bug in your configuration!
Followed advice and found my configuration folder (same as visible by samba share) and created meter files with sudo nano and IT WORKS
Looks like the problem were files created within windows notepad same data saved with sudo nano stated for victory
THANKS CPLANT
Hehe is working config for TheengsGateway - it is BT gateway working on Rpi 4 (maybe other) without additional HW.
theengsgateway:
image: theengs/gateway:latest
container_name: theengsgateway
network_mode: host
environment:
MQTT_HOST: 192.168.1.247
MQTT_USERNAME: mosquitto
MQTT_PASSWORD: k0mar23
MQTT_PUB_TOPIC: home/TheengsGateway/BTtoMQTT
MQTT_SUB_TOPIC: home/TheengsGateway/commands
PUBLISH_ALL: true
TIME_BETWEEN: 60
SCAN_TIME: 60
LOG_LEVEL: DEBUG
HASS_DISCOVERY: false
DISCOVERY: true
DISCOVERY_TOPIC: homeassistant/sensor
DISCOVERY_DEVICE_NAME: TheengsGateway
DISCOVERY_FILTER: "[IBEACON,GAEN,MS-CDP]"
SCANNING_MODE: active
ADAPTER: hci0
volumes:
- /var/run/dbus:/var/run/dbus
volumes:
data_theengsgateway_data:

Followed advice and found my configuration folder (same as visible by samba share) and created meter files with sudo nano and IT WORKS
Looks like the problem were files created within windows notepadsame data saved with sudo nano stated for victory
THANKS CPLANT
Thanks, glad to hear. Finally. Will still have to investigate the wmbusmeters-volumes-topic. Am not sure whether I currently have everything backed up correctly. Having three volumes on my end for the same thing makes me kind of nervous.
Well done, thank you for great tutorial!
My few questions and comments:
- As I was migrating from instance where I already got influx2 I got stuck as there is no influx2 for armv7. Hopefully Iām on RPi4b 8GB so I just went for 64bit lite version and everything seems to be working
- Any reason why you are using full paths when mounting to certain services (chronograf, duplicati) instead of volume?
- BTW it creates potential issue if someone lables the master project differenty (as I did) as the full path has ā¦/openhab_data⦠hardcoded in
- also sometimes the timezone is parameter (eg. openhab), sometimes its hardcoded (eg. duplicati, samba)

As I was migrating from instance where I already got influx2 I got stuck as there is no influx2 for armv7. Hopefully Iām on RPi4b 8GB so I just went for 64bit lite version and everything seems to be working
Youāre right, I noticed this as well. The influxDB-part is one of the things that needs some thought going forward, as 1.8 will come to an end sooner or later.

Any reason why you are using full paths when mounting to certain services (chronograf, duplicati) instead of volume?
With precision youāre pointing to a part which I was always happy about that itās working, but which I solved probably in a not-so-ideal way. So Iām happy to incorporate any changes or recommendations which you have. Feel free to post what youāre running, and Iām happy to replace it above.

BTW it creates potential issue if someone lables the master project differenty (as I did) as the full path has ā¦/openhab_data⦠hardcoded in
Same here. As mentioned in my disclaimer, Iām not a pro. If you let me know how you would solve it, and if it works, Iām happy to edit the instruction above.

also sometimes the timezone is parameter (eg. openhab), sometimes its hardcoded (eg. duplicati, samba)
Same as above. Happy to incorporate whatever suggestion you have.
Thanks, Iām currently working on the setup, Iāll post the outcome when Iām more happy with the results⦠Iām still getting some issues (dangling volumes from samba, wrong userids, unaccessible folders), but moving slowly forward as a night-time hobby ;-). Iām not as good as local gurus (hats off, Rich), but also thought of learning (reusing/tailoring) ansible to automate some of the steps to make it easier. If someone needs help sooner, I can try (most things currently work).

hats off, Rich
Couldnāt sum it up any better. Started here around two years ago, and still feel somewhat stupid, even though the tutorials that I wrote are probably not that bad.
Anyway, as soon as you have something, let me know. We could also go ahead and move the entire docker-compose.yml to Github.

I added zigbee2mqtt to the stack
I just got a Zigbee device and considered to go the zigbee2mqtt-route. Instead I just installed the Zigbee-Binding, which works like a charm as well.
This is a great setup! Thank you very much!
I have to change it a bit to work for my needs. I like the setup where the persistent data from the dockers is saved in the host file system, like it is done in the openhab/docker-documentation.
I donāt need duplicati this way, and also I donāt need
But I am struggling to get it right for frontail:
can't open '/var/log/openhab/events.log': No such file or directory
can't open '/var/log/openhab/openhab.log': No such file or directory
I think I should change the location of the logfile to where it is on my host, but how? I changed it like this, but itās not working.
volumes:
#- /var/lib/docker/volumes/openhab_data_openhab_userdata/_data:/openhab/userdata:ro
- /var/lib/docker/volumes/openhab_data_openhab_userdata/_data:/opt/openhab/userdata/logs:ro
And also I wonder how to set this correctly for influxdb and grafana too. Do I have to make users āinfluxdbā and āGraphanaā and set the user-idās just the way it is handled for openhab?
Unfortunately not able to answer your question. Tbh, I was pretty happy to get everything working the way I wanted and put it into this step-by-step (especially on the volumes I have to admit that I have not fully understood this topic), so I hope thereās someone else out there with more basic understanding.