Page 1 of 2
zwavejs-ui - FAIL - how many of you did try & succeed?
Posted: Wednesday 14 February 2024 18:25
by renerene
OpenZwave -> zwavejs-ui
after two days work, I gave in. Went back to my old domoticz version.
Please tell me about your attempts. Just to get an idea of succes rate.
I used the latest 2024 Dz version and followed Wiki carefully. There's something wrong, new devices are not registered correct and DZ keeps finding the same devices every 2 seconds
Re: zwavejs-ui - FAIL - how many of you did try & succeed?
Posted: Thursday 15 February 2024 0:31
by waltervl
Read this topic.
viewtopic.php?t=39292
It is not easy as you have to invest some time in learning and understanding new applications like MQTT, Docker and of course Zwave-JS-UI.
But hey, it is an hobby else you better buy a commercial smart home system like a Homey.
Re: zwavejs-ui - FAIL - how many of you did try & succeed?
Posted: Thursday 15 February 2024 9:38
by BartSr
@renerene
Yes, it's a steep learning curve but my experience learned that if you carefully follow the instructions as pointed to by Waltervl
https://www.domoticz.com/wiki/Zwave-JS-UI you will be able to use Zwave in Domoticz.
Can you tell what you tried so far?
MQTT already running on your system? In that case you can skip the part about MQTT and it's just a matter of configure ZwaveJs.
Than only install Docker is left and creating right yml file. And ofcourse configure ZwaveJs (which is not that difficult once you understand).
-Bart
Re: zwavejs-ui - FAIL - how many of you did try & succeed?
Posted: Thursday 15 February 2024 11:04
by lost
waltervl wrote: ↑Thursday 15 February 2024 0:31
But hey, it is an hobby else you better buy a commercial smart home system like a Homey.
Not sure this is the right answer. The main issue is with zwaveJS base design choices that brings a dependency burden unique to this radio/protocol management.
That's so unbelievable you may think about some strategy behind this: HA is installed by most people (that do not already buy HA hardware, the easy Homey way!) as a full image so that's easier/hiding the burden, but it really hurts the way other DIY home management solutions worked.
So if I had a plan (let's call it "HA to rule them all"), being based on free SW, making custom installs as difficult for users (and breaking former APIs for designers) as I could would be an objective to achieve some next steps.
Re: zwavejs-ui - FAIL - how many of you did try & succeed?
Posted: Thursday 15 February 2024 13:33
by waltervl
lost wrote: ↑Thursday 15 February 2024 11:04
waltervl wrote: ↑Thursday 15 February 2024 0:31
But hey, it is an hobby else you better buy a commercial smart home system like a Homey.
Not sure this is the right answer. The main issue is with zwaveJS base design choices that brings a dependency burden unique to this radio/protocol management.
There is nothing Domoticz can do with this. zwavejs also now enables the same Domoticz implementation method as Zigbee and other protocols through MQTT AD Gateway. For users with different protocols it will be easier.
So for those declining users of Zwave it has become more difficult but there is no other way. Or you want to improve Openzwave again. Lots of users are telling ZwaveJSUI performs much better then the Openzwave implementation so after the initial installing issue there are more happy users.
As Johan Cruijff says: Every advantage has his disadvantage.......
Re: zwavejs-ui - FAIL - how many of you did try & succeed?
Posted: Thursday 15 February 2024 21:26
by akamming
lost wrote: ↑Thursday 15 February 2024 11:04
HA is installed by most people (that do not already buy HA hardware, the easy Homey way!) as a full image so that's easier/hiding the burden, but it really hurts the way other DIY home management solutions worked.
I think you are oversimplifying how things with HA work. Just for fun i am now also experimenting with HA. I found out HA also uses Zwave JS UI. So if you cannot convert your openzwave to zwave js ui in domoticz, you also won't succeed with HA...
The interface and first installation is indeed simpler, but then again other things are really complex in HA (e.g. i have 2 simple dzvents scripts for which i have not yet found a way to convert to HA and on HA forum no one has been able to provide me the solution yet)
Have fun trying HA, but be careful with suggesting everything in HA is better than domoticz. It is not my experience. And if you can't get ZWAVe JS UI to work, you will have the same issue in HA
The grass looks is not always greener on the other side
Re: zwavejs-ui - FAIL - how many of you did try & succeed?
Posted: Friday 16 February 2024 0:36
by solarboy
Succeeded when Auto Discovery first arrived. Thanks to the instructions I had very few issues, a few devices couldn't be "replaced" as they were slightly different, for example RGB lamps but otherwise no issues and the benefits vs OpenZw were sooo good.
Re: zwavejs-ui - FAIL - how many of you did try & succeed?
Posted: Friday 16 February 2024 12:36
by lost
akamming wrote: ↑Thursday 15 February 2024 21:26
The grass looks is not always greener on the other side
Sorry if this was not clear, but that's not in my plans to go HA. I'm really happy with Domoticz since more than 7 years now!
Was just thinking HA may have future plans to be more commercial (they already sell HW, like Jeedom) and taking the opportunity to drag such a common home automation protocol, then it's users with something so labyrinthine to handle out of my system, would make sense. Hope I'm wrong, but seeing MQTT-AD also announced step back really rings a bell...
Re: zwavejs-ui - FAIL - how many of you did try & succeed?
Posted: Friday 16 February 2024 15:55
by BartSr
@lost
Can you be more specific about step back of MQTT- AD?
Re: zwavejs-ui - FAIL - how many of you did try & succeed?
Posted: Friday 16 February 2024 16:58
by akamming
lost wrote: ↑Friday 16 February 2024 12:36
akamming wrote: ↑Thursday 15 February 2024 21:26
The grass looks is not always greener on the other side
Sorry if this was not clear, but that's not in my plans to go HA. I'm really happy with Domoticz since more than 7 years now!
Was just thinking HA may have future plans to be more commercial (they already sell HW, like Jeedom) and taking the opportunity to drag such a common home automation protocol, then it's users with something so labyrinthine to handle out of my system, would make sense. Hope I'm wrong, but seeing MQTT-AD also announced step back really rings a bell...
They are just stepping forward on ZWAVE JS:
https://www.home-assistant.io/blog/2024 ... -alliance/. Which is good news for ZWAVE JS users on domoticz as well... They are preferring the websocket layer however, so there is a bit of risk on the mqtt layer for zwave js. Fortunately the developer stated lately he is not planning on removing that feature, cause a lot of happy users (including ha users) are using the mqtt layer...
i like it as well.. the integration is much more transparent. (e.g. now both my domoticz and ha instance control the same zwave network, which is quite nice...)
Re: zwavejs-ui - FAIL - how many of you did try & succeed?
Posted: Saturday 17 February 2024 8:17
by lost
BartSr wrote: ↑Friday 16 February 2024 15:55
@lost
Can you be more specific about step back of MQTT- AD?
Hello,
Read this & make your own mind:
https://github.com/domoticz/domoticz/issues/6021
Re: zwavejs-ui - FAIL - how many of you did try & succeed?
Posted: Monday 19 February 2024 21:38
by Tuinfluiter
When I first read about the migration to Zwavejs-UI and read the instruction, I thought: I'm not going to do this.
The second time when I read it, I thought the same.
I installed HA for sometime on a spare Pi: installation: great; integrations: OK, also some learning curve, but doable. I didn't install any hardware; just the buienradar. When I looked at the data and database that was collected out of the box, I thought: this can't be true. No aggregation of data (not that I noticed or could find). After a month I pulled the plug out of HA.
The data that Domoticz gives me right away is intuitively what I want.
So back to the Zwave migration. I collected a lot of courage, started preparing, and got started.
I ran into a lot of troubles. I didn't have experience with docker. In total there are so many variables; so many places where something can go wrong; it all happened.
Some of the problems: Buster as an "old OS version"; problem with authorization; misunderstanding of the security keys of Zwave; at first didn't notice the "save" button in Zwavejs-UI (it's the type of UI as on a mobile phone, where I don't expect a save button), localhost vs IP adress (thx waltervl); zwave stick not connected (fuzzy message about client not connected, oh of course the client is the stick, of course); auto discovery: some screen shots show a setup button; others don't show it.
There are simply too many variables where things can go wrong.
But I succeded. The nice thing is: OZW can be disabled and enabled, so you can keep the hardware running untill the moment you're ready and switch definitely to Zwavejs-UI.
I think it took me some 1.5 weeks of lead time; very, very, very many hours of effort.
Re: zwavejs-ui - FAIL - how many of you did try & succeed?
Posted: Monday 04 March 2024 11:57
by DaveWils
I tried and keep failing, just can not access the ZWAVE-JS UI web interface to configure anything. Costing me hours. it does not give an error on startup, there's no logfile...
Re: zwavejs-ui - FAIL - how many of you did try & succeed?
Posted: Monday 04 March 2024 12:50
by BartSr
Can you provide more details. The info sofar does only say you spent (too) much time and you didnot got zwavejs running.
pls provide
-docker compose template content
result of linuxcommand:
sudo ls /dev/serial/by-id That is reverting data to be used in the container template.
Bart
Re: zwavejs-ui - FAIL - how many of you did try & succeed?
Posted: Tuesday 05 March 2024 21:33
by DaveWils
Re: zwavejs-ui - FAIL - how many of you did try & succeed?
Posted: Tuesday 05 March 2024 22:20
by tjabas
Is it possible to get zwavejs-ui integrated in domoticz in the future somehow? So it comes with a new installation of domoticz
Re: zwavejs-ui - FAIL - how many of you did try & succeed?
Posted: Tuesday 05 March 2024 22:22
by waltervl
That will probably never happen as it are 2 different applications.
Re: zwavejs-ui - FAIL - how many of you did try & succeed?
Posted: Sunday 17 March 2024 14:50
by DaveWils
Well, if this docker thing is da shit, why can't it install 2 (or 3) applications?
My heating hasn't been working since the january new year surprise update with no openzwave.
No longer needing heating now, so can get through summer and then move to a new system (IoBroker, Home Assistant), whoever has an installation image, like domoticz used to have, in fall.
Cannot get ZWAVE-JS-UI to work and even if that ever works I still need to replace all devices one by one, might as well bite through then to the system that has the most active developers in fall...
Re: zwavejs-ui - FAIL - how many of you did try & succeed?
Posted: Sunday 17 March 2024 14:56
by madpatrick
DaveWils wrote: ↑Sunday 17 March 2024 14:50
My heating hasn't been working since the january new year surprise update with no openzwave.
OZW has been back since version 2024.2.
Re: zwavejs-ui - FAIL - how many of you did try & succeed?
Posted: Saturday 08 June 2024 22:31
by riom16
Hi,
2 Raspberry Pi 3 family.
First one : 2 afternoon and it's working. With minor issues.
Second one : Stuck on bug
https://github.com/zwave-js/zwave-js-ui/issues/3639.
It was so simple and so reliable in openzwave