Testing Z-Wave binding on openHAB-2

Yes - it’s highly likely. I expect it’s the same problem we came up against yesterday with the network security setting.

I can stop the error by trapping the exception (Assuming it’s the same exception we had yesterday), but if the system won’t allow changing if configuration for things configured through text files, then this is quite a big problem that I have no answer for. I can send the updates to the device still, but as the UI won’t change, you will have no feedback that anything is happening.

I was afraid of that!
Well, at least I then I know. Have to think about how to proceed from here though. I really like the textual route though, so I guess I’ll just hang in there for a while, and see if there’s any decisions in either direction regarding this.

I’m not familiar with how this works under the hood, but I would hope that ultimately, configuration can be stored in the mapdb even for things/items defined in text files…

Yes, this would be great, or some way of exporting mapdb into something editable/readable. Import/export might be good enough.
If I eventually need to go the full mapdb route, looking at it from a historical view, I suppose that I would have to be prepared to build up a new mapdb from scratch every now and then? Or did the mapdb format stay the same for a long time?

I don’t think so. The problem with this is you’ll send a config change through HABmin, and nothing will change in HABmin if it’s using static data. The binding can still update the device config, but unless configuration is treated dynamically, I think it will be a mess.

I don’t expect it to change.

I meant from a users upgrade perspective - statically export a full mapdb, [then adapting to a new format] followed by importing to a new, fresh mapdb. Then all “runtime” changes are managed through the UI’s. I might be wrong, but currently mapdb from a users (e.g. me :slight_smile: ) perspective is just a big black hole. This is of course then outside the zwave binding.

I suppose most people does not have too many things in OH2 yet, but populating my setup properly from scratch I’m guessing at least a full week-end, probably more, even if I save notes about which nodes needs which configuration. And worse; very tedious of course. :slight_smile:

I know there’s another discussion around this, somewhere, and I realize the problem marrying textual an GUI settings. Maybe the best compromise is what you are suggesting, with textual items having a overlay in mapdb. At least I guess it might not need any radical changes.

I can email you a full log … :slight_smile:

however that node looks not so good:

22:05:03.518 [ERROR] [curityCommandClassWithInitialization] - NODE 17: Invalid state! secure inclusion has not completed and we are not in inclusion mode, aborting
22:05:03.523 [DEBUG] [curityCommandClassWithInitialization] - NODE 17: call from NodeAdvancer initialize, from xml, handing back message=null
22:05:03.526 [DEBUG] [nitialization.ZWaveNodeStageAdvancer] - NODE 17: Since secure inclusion failed, the node must be manually excluded via habmin

so I sent you 2 mails:

ZW100 length error I get
and
Devolo Motion Sensor stays unknown

Thanks! :smiley:

EDIT: after multiple tries… the Devolo now received the correct name in Habmin! so the log I sent probably shows only a correct initialization. However I had to delete the thing and just restart discovery once more then it was pulled correct.

Same thing I did before quite a couple times…

ADDITIONALLY a NODE1 ! was found now… that should be my controller?! no clue why it shows up in discovery in habmin now???

regarding editing stuff in the DB

for the devolo http://www.cd-jackson.com/index.php/zwave/zwave-device-database/zwave-device-list/devicesummary/290

  1. I would like to delete the Options from the database (parameters 2,3,4)
    No possibility to do so :slight_smile:

  2. also the Command Classes are not ok (they are populated from the XML?)
    e.g. there is DOOR_LOCK in there which is not available for the PIR sensor
    delete these non correct classes seems also not available to me?

thanks

I guess you need to do what it says :slight_smile:. It looks like secure inclusion failed - this code needs some work still…

There are two places where command classes come from - a list in the binding (for the device classes, there are - in theory - a set of mandatory command classes), and the device itself (and there’s two places where the device provides this list). I’ll take a look at the log you sent…

I’ll delete the parameters… I’ll also try and work out the copy function tomorrow… (sorry that’s taken some time).

OK that explains that.
Where does the device class come from in first place?
This one is says SECURE_KEYPAD_DOOR_LOCK for the devolo… Which is wrong … It’s a pir + Temp \ Lumi

The device.

Remember that this is all reverse engineered, so occasionally we’ll get things wrong, but we have to make an educated guess from other devices… I’ll have another look at this when I get a chance…

@shorty707 this should be fixed. Let me know if you have any more problems…

Why? I just looked at parameter 2, and it looks ok - doesn’t it?

Hi,

just the Options part of these parameters. (2, 3, 4)

I understood that, but my question is what is wrong with this. If I look at your example above, we have option 0=Off, and it says exactly this in the Overview - same for 255=On. I’m not sure what is wrong with the definitions?

I thought the1-100 in the overview might be a wanted setting for some users?

Of course, you can write this sort of thing in the overview as well - that’s fine.

I saw you updated the device… just to understand

when there are Options. These are the only possible setting I can do for that parameter.

Eg. Parameter 3 --> PIR Sensitivity 1-99

there is still a “Option” which will block me from entering any value from 1-99 I would like to enter