Migrating from homeassitant: zwave devices unknown

I’ve recently migrated from homeassistant. Most of my devices appeared in zwave binding ok. 3 are having trouble

  1. MCOHome 3way touch switch MH-513 (Unknown device 015F:5131:5103:5.0)
  2. MCOHome 3way touch switch MH-513 (Unknown device)
  3. Schlage Link Mini Keypad RFID (Unknown device)

I don’t know why what I though two identical devices would not both identify the same.

The Schlage link identified from HA/openzwave as;
|Manufacturer ID:|0097|
|Manufacturer:|Schlage Link|
|Product:|Mini Keypad RFID|
|Product ID:|5501|
|Product Type:|6131|

I’ve registered for zwave binding online database and awaiting edit rights. I’ll follow up here as I go.

Is there a sensible translation between HA/openzwave xml files and what I need for the zwave binding database entry?

Not that I’m aware. The easiest way to initially populate the database is by uploading the node.xml file that’s created in the userdata/zwave directory. That will save a great deal of work.

And welcome to the world of openHAB!


I have had some success with two of my things, thanks to the binding database updates.

However, one of my MH-s513 refuses to have its manufacturer identification detected. To me the 2 MCOHOME switches I have look identical and both worked without problems in homeassistant. I have excluded and included this device many times to no avail.

I have enabled debug mode in the binding, attaching my log, then node in question is pesky Node49.

log.txt (268.8 KB)

Can you provide the log from a little earlier please?

The issue is that the binding is requesting information about a command class (ie the version of the multi-channel CC) and the device is not responding. Probably the binding is doing the as the device reports it supports this CC, but I need to see the inclusion response to confirm this.

I’m not completely sure how to handle this if this is the case as this is very early in the initialisation process, and we don’t know what the device is at this point, so the database can not be used to perform such configuration.

Here is a log with a fresh include to Node50 after excluding Node49.

I appreciate you looking into this.

log.txt (108.5 KB)


So the real issue here is that the device is not responding to the request for the manufacturer data -:

We see the binding send the request for the manufacturer information, and the device sends the ACK to say that it received the command, but we never receive the data. This repeats continuously.

I’m not sure what to suggest with this - the previous log showed a similar sort of response (ie. no response) to the VERSION command class, so something very strange is happening with this devices responses.

The only thing I can suggest is to reset the device to factory defaults to see if that helps. In theory, excluding the device should do the same thing, so I’m not convinced it will help, but it’s about the only thing I can think of at the moment.

TL;DR “Did you turn it off and on again?”

I tried many ways to get this device up and running;

  • hard reset the controller and added only the two switches,

  • install fresh openhab 2.3 stable

  • install fresh homeassistant

In no configuration of the above did the offending switch ever send back a response to a query. I did through some xml hackery get the switch to look like it was working in homeassistant, and yes I could turn it on via UI, but it would never report its state so I could not turn it off >.<

It turns out that this mains powered zwave switch needed a reboot via the circuit breaker. I think credit for this goes to a hint from @JjS I would have thought factory reset would be enough…

Now a fresh openhab 2.4 M8-1 discovered both switch panels and things seem to be playing nice. I will report back once I re-include all the rest of my gizmos.

Good catch!

I see this occasionally from the Leviton switches and dimmers. The dimmers are easy due to the built-in air gap switch. The switches usually require a power cycle at the circuit breaker…