Node:undefined undefined

[tomdh76 (Tom den Heijer)]

I also suddenly had all my nodes at node:undefined. In my case it could revive it by restarting my virtual machine on which I have a docker HASS…pffff

[gohassgo]

@stephenmarcantonio

This is possibly the same issue as https://community.home-assistant.io/t/moved-ha-and-lost-zwave-mostly/166860 , which can be fixed (temporarily?) by restoring an old version of your zwcfg_*.xml file (if you have a backup).

I replaced zwcfg_*.xml from a backup of 0.107.6 and restarted and all zwave entities/devices came back. Thanks!

[gderber]

I had the same issue myself today. Unfortunately, I added the device late last night and hadn’t backed up yet.

I was lucky, I had another device of the exact same type (Both Aeotec Recessed door sensors). I copied the section <Manufacturer …> to from the working sensor to the undefined sensor. The undefined sensor started working again. At least well enough for me to do a ‘heal node’ / ‘refresh node’

So in my case:

<Manufacturer id="86" name="AEON Labs">
     <Product type="102" id="59" name="ZW089 Recessed Door Sensor Gen5" />
</Manufacturer>

Update: To get the entities back afterwards, I had to copy the entire Command Classes section as well.

<CommandClasses>
...
</CommandClasses>

[ibennani (Ilias Bennani)]

Running 0.107.5 and had the same issue.

Solved it by selecting that node, tapping the Node info-button and there I found an option to remove that node.

Now it’s gone from my system.

[eddriesen]

hassio on Ubuntu 18.O4, had 30% of my nodes dissapear after upgrading to 0.107.5 , shutting down, replacing zwc_*.xml with one out of a snapshot (extract the previous snapshot /usr/share/hassio/backup/xxxxxxx.tar and cherrypick zwx_*xml out of homeassistant.tar.gz) fixed it. Thanks guys!

EDIT:Did something change in the zwave stack. I have no numbers, but loading network seems waaaaay slower.

[shadgkiss (Steven Hadgkiss)]

I also am having this issue, just updated to Core 110.6 and supervisor 227. Zwave has been extremely painful for the last few versions since last week. Of course I dont have a backup of the file I need.

I am seeing this:
Aeotec Limited ZWA005 TriSensor (Node:undefined undefined)
Unknown Node 2 (Node:2 Complete)
Unknown Node 5 (Node:5 Probe)

[rzen (Rouslan Z)]

chiming in with an unhappy “me too” welp.

in my case its coincided with a network upgrade and a move to a different host. my HA is a vanilla docker setup. about 30-40% of nodes are “undefined” and grayed out. no node info available:

alas i do not have a recent enough backup of zwcfg… i have 40+ devices, hundreds of customized entities… this isnt going to be fun unless there’s a solution.

a side question, i’m on aeotec gen5 z-stick now. anyone knows of an alternative with linux-friendly tooling for backup/restore? thanks.

[rzen (Rouslan Z)]

While browsing around i’ve come across this post:

Moved HA and lost Zwave mostly

This happened to me, and it looks like something caused my zwcfg_*.xml file to be corrupted with a lot of important values replaced with zeros and defaults. For example, these lines in the working file… <Node id="8" name="" location="" basic="4" generic="33" specific="1" roletype="5" devicetype="3079" nodetype="0" type="Routing Multilevel Sensor" listening="true" frequentListening="false" beaming="true" routing="true" max_baud_rate="40000" version="4" secured="true" query_stage="Complete"> <…

Which looks a lot like what happened in my case. It might be a clue as to what is happening here.

Like i mentioned, in my case the troubles coincided with a network upgrade which entailed quite a few incidents of intermittent network unavailability. Perhaps unluckily timed network outs/lags cause OZW to corrupt zwave network’s internal configuration/state somehow and ultimately dump it into zwcfg_ file. I wish i knew more about inner workings of OZW to be more specific.

Oh and i ended up finding a slightly stale backup for my zwcfg, not perfect but way better than re-including 30 devices (whew!).

[Dingin (LEB)]

same issue here

[Mil]

I have the issue here also. Only 2 devices… a sensor & a lock. Yesterday I tried excluding then including one of them which fixed the problem temporarily, but now it shows undefined again.

[uoah (Ferran)]

Hi all,

At least I’m not alone… I have been running homeassistant perfectly configured and some days ago when I tried to add some new devices it started the nightmare. The thing is that I succesfully recovered some nodes but when I recover some of them I lose other ones, I try to add again and I lose some already recovered and sometimes I cant recover anything…

It’s very frustrating because you don’t know how much you use some automations until you lose them…
I hope we find a solution soon because I have been dealing with this issue this last three weeks.

[paulstronaut (Paul)]

This happened to me today. I’m not ready to switch to the OZW add-on and was worried that I’d have to jump into it. I tried many things like restarting the zwave network, restarting hassio. In the end, it took a few restarts of the hassio OS itself before all nodes finally started registering correctly.

[MarcellusWallace]

Hi All,

I’ve read many articles but none is actually providing a solution.
I have one device that wasn’t added well and now it shows like

Unknown Node 22 (Node:undefined undefined)

[iMiMx (iMx)]

Just had the same, after a power cut and UPS depleted, was running 2021.5.1. fix for me:

• On the CLI stop ha core, this is important before replacing the file

ha core stop

• Replace zwcfg*.xml with on from a snapback (Download snapshot, extract, extact homeassistant.tar.gz. I replaced with a version from 2021.4.6

• Start ha core again

ha core start

I believe if you replace zwcfg*.xml with HA running, it replaces/save on restart.