Devices going "unavailable" while still being green in Z-Wave JS UI Edit: 10 branch

Configuration Z-Wave
1

I have a bunch of dimmers of the same brand that go unavailable at random times in Hass, but it does not seem to go dead in Z-Wave JS UI, it’s green whenever I check it there. I am trying to make an automation to ping them when it happens, but I’m not sure if it works because I’m not sure that unavailable is “Dead”, which is the alternative for Z-Wave. For Zigbee “Unavailable” is a choice, and that works perfectly for my Easy Access door lock, which falls out every two weeks or so and has to have the batteries removed to work again (this is unrelated to my system, it’s a fault in the Zigbee chip for that device - I have had threed different Zigbee chips for it, and I have tried it with three different Zigbee controllers in the Pi). This is what the log book looks like for “Badedimmer”, wich is Bathroom dimmer:

image
image1633×3519 507 KB

Is “Dead” the same as “Unavailable” in Zigbee, or do I need to look for another property? Here is the automation:

alias: Pinge badedimmer når den blir utilgjengelig
description: ""
triggers:
  - device_id: 0d29df3e5bfb09eb56c294775a51e505
    domain: zwave_js
    type: state.node_status
    entity_id: 8e52911ec3884ba44dee9d26f090bbf7
    trigger: device
    to: dead
conditions: []
actions:
  - action: notify.automatiseringsepost
    metadata: {}
    data:
      message: Der falt badedimmeren ut
  - action: zwave_js.ping
    metadata: {}
    data:
      device_id:
        - 0d29df3e5bfb09eb56c294775a51e505
mode: single

But of course finding the root course would be better. Does anybody have an idea why this is happening? They all go unavailable at the same time, it seems. It often is related to restart of Hass, but not only that, they can go unavailable at any time. I wonder if it could be because they are not sending alive messages, or at least not sending them often enough?

Edit: I have now found that it affects one of the motion sensors too. There is something seriously strange here. I have rebooted the Pi, and I have pulled the latest Docker containers of everything, but still this is happening. I have not had this problem before, as far as I know.

Edit 2: I think I have found the problem. I downgraded from the latest version to 9.33.0.92d89c5, and that seems to work as it should.