Z-Wave graph (without the python)

My html file said Version 2.3: (03 February 2020) - I updated to the May version as well with no luck.

In the end, I got it to work my manually specifying the webcomponent_path in my configuration.yaml.

Now, I’m trying to figure out why everything is black!

(Chrome 83.0.4103.106 on Mac)

Update: black boxes only occur when going through my nginx proxy. When accessed via the local url, it works fine.


Any idea why my graph isn’t linking all the goodies together?

4 of the 6 unconnected nodes are battery powered. When is the last time they were awake to report in?
Of the 2 remaining nodes, 1 is unknown/not configured so I can’t say, the last 1 is a light bulb and also showing unconnected. Is it powered off at the switch? The graph can’t map unconnected devices.

Tank you that makes sense. Since I posted this there have been more things showing connected. Still showing the Inovelli bulb unconnected which has been working just fine from a Z-Wave perspective.

The unknown node is a mystery to me - it showed up when I first set up Z-Wave and I can’t seem to figure out what it is.

Just a heads up guys:

Specifying custom panels based on HTML imports is deprecated and will be removed in a future version.

yes, received this:

and check: https://www.home-assistant.io/integrations/panel_custom#webcomponent_path
and breaking changes: https://rc.home-assistant.io/blog/2020/06/24/release-112/#breaking-changes

now how to proceed ?
still working for now, but would be nice if we could rebuild this to be future proof

odd thing is I am using:

panel_custom:
  - name: zwavegraph2
    sidebar_title: Z-Wave Graph
    sidebar_icon: mdi:access-point-network
    url_path: zwave

which doesn’t use the documented deprecated webcomponent_path

If you use the ozw beta, you can’t use this graph right now lol

1 Like

It still works fine. that’s not it. I was pointing out some of the important referencing posts and docs, so we could try to mitigate the upcoming deprecation.
hope @NigelL is still around to see if he can update his work, and of course @AdamNaj ,please forgive the tags, I didnt see any other way of pinging you here.

I think it shows due to the content of the files just being html or having the .html extension :man_shrugging:

It 100% doesn’t work with the new container. The required entities do not exist.

Don’t understand then, maybe I see it wrong, but I still have the ZWave test graph:

No clue. I can’t say why it’s working for you. It works off the zwave.xxx entities which the new openzwave integration does not do create.

must confess I am not sure what I use is the Open Wave integration you mention, but I use this:

the built-in Z-Wave integration and an Aeon Labs usb stick creating these entities:

Ok, that’s not the openzwave beta. That’s the old zwave setup.

EDIT: All the 3 zwave setups in home assistant use openzwave. Hence the confusion. You are using the built in zwave setup which is using openzwave 1.4 (a 3 year old build of openzwave).

ok thanks…
anyways, the deprecation of the html file use has been announced, so I hope the author will find time to update that.

1 Like

This addon will likely need a complete rewrite once Z-wave is decoupled from Home Assistant in the near future. If that is of concern, I’d recommend separating your current Z-wave environment into its own separate HA install that can remain in stasis until you’re ready to migrate to the new system.

well, I already am using a dedicated Zwave instance :wink: and a dedicated Mqtt broker…

It’s not that I see he graph as a critical part, I’ve only installed is because I could, and now appreciated the representation of the nodes.

But, are you saying the built-in Z-wave add-on will be abandoned? Couldn’t imagine the team to do so tbh. What do you mean with being ‘decoupled’ ?

While the official word is the existing Z-Wave integration isn’t going anywhere, if this new method works out as planned, it’s a logical conclusion that the current integration will be going away. It’s already an established pattern.

Step 1 - Devs create a new way of doing something that already exists while reassuring the masses that the current way is staying put.
Step 2 - The new way becomes so appealing that the masses gleefully begin adoption.
Step 3 - Deprecate the old because everybody is happy and old is dumb.

In the long run, the only way the current Z-Wave integration remains is if the new method fails horribly. I don’t see that happening as there are extremely intelligent people working on this new feature.

To be fair the upstream of OZW 1.4 is no longer maintained so there will be no fixes coming out for it or any updates that are done to the zwave standard/protocol, such is the life cycle of software.