Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • Dec 31 2020 11:33
    uuperi commented #322
  • Dec 22 2020 16:03
    robertsLando commented #393
  • Dec 22 2020 16:02
    robertsLando commented #393
  • Dec 22 2020 14:28
    nouknouk commented #393
  • Dec 22 2020 14:28
    nouknouk commented #393
  • Dec 22 2020 14:27
    nouknouk commented #393
  • Dec 22 2020 14:23
    nouknouk commented #393
  • Dec 22 2020 14:06
    robertsLando commented #393
  • Dec 22 2020 14:01
    nouknouk commented #393
  • Dec 22 2020 14:01
    nouknouk commented #393
  • Dec 22 2020 12:58
    robertsLando commented #393
  • Dec 22 2020 12:57
    robertsLando commented #393
  • Dec 22 2020 12:44
    nouknouk commented #393
  • Dec 22 2020 12:40
    nouknouk commented #393
  • Dec 22 2020 12:28
    robertsLando commented #393
  • Dec 22 2020 10:22
    nouknouk edited #393
  • Dec 22 2020 10:21
    nouknouk edited #393
  • Dec 22 2020 10:19
    nouknouk opened #393
  • Nov 16 2020 17:36
    GetUpdate commented #274
  • Nov 12 2020 17:05
    robertsLando commented #392
Jordan Green
@jordan-green
Or are you saying we need to take that information in and manually write it to the device via serial port ?
Elias Karakoulakis
@ekarak
its quite possible your zwave device isn't known by OpenZWave
usually there's a heapload of xml files under /etc/openzwave
try locating your device there
also try looking at the github repo - some other user might have added it
Jordan Green
@jordan-green
It definitely is - it's an aeotec
Is there any chance you have an example snippet of the initialisation then storage of config you could share ?
I've got everything else working and basic commands working, but manufacturer specific commands etc pose a problem as no device type of manufacturer id is available - on adding the devices these appear blank
Elias Karakoulakis
@ekarak
OpenZWave/node-red-contrib-openzwave@b75a5ae
is an example of writing out the config file , which you should call after you add new nodes into your controller.
Jordan Green
@jordan-green
Thank you very much @ekarak will let you know how we go
Sebastian Raff
@hobbyquaker

Hi! I'm trying to use a Fibaro Button with node-red-contrib-openzwave and can't get over Warning, Node008, No ValueID created for Scene 1. Used createScene and addSceneValue, my zwscenes.xml looks like this now:

<Scenes xmlns="http://code.google.com/p/open-zwave/" version="1">
    <Scene id="1" label="Scene 1">
        <Value homeId="0xdd2f894a" nodeId="8" genre="system" commandClassId="115" instance="1" index="1" type="byte">1</Value>
    </Scene>
</Scenes>

Can someone give me a hint what I'm missing to get the scene events? Is createScene/addSceneValue enough or do I have to do more?

Tony Tam
@fehguy
Hi folks, this is a great project. I was using zwave.me for a while but got tired of how flaky it was and have switched a 60 device network over to node-openzwave-shared. It works almost flawlessly except after a few days of usage, the zwave instance will receive a notification state of timeout and those devices are never reachable again until I restart the zwave software
I have tried the zwave.healNetwork() command, as well as zwave.requestNodeNeighborUpdate(deviceId) but no such luck
any tips would be greatly appreciated!
Elias Karakoulakis
@ekarak
what you're describing is a controller lock-up
it's a bug in the underlying C++ library
Tony Tam
@fehguy
Hi @ekarak thanks for the reply--any tips for working around it?
Elias Karakoulakis
@ekarak
which OpenZWave version are you using? you might try upgrading the OZW lib
Tony Tam
@fehguy
let me see...
master :) . Had a different issue with 1.4
Elias Karakoulakis
@ekarak
let me see on my side - I vaguely recall having the same issue some time ago
Tony Tam
@fehguy
i have a number of devices being polled. It's possible that I can reduce the number, if that would help
Elias Karakoulakis
@ekarak
bah, its stuck on my own raspberry too
Tony Tam
@fehguy
:D
I've written some code to detect this state, I could (cough cough) restart things when they get bound up. but that's ugly
Elias Karakoulakis
@ekarak
OpenZWave 1.4.2818
ok - the issue here is that according to the guys at the core lib the serial controller on the chip can lock up, so you'd need to do some pretty low-level stuff at the USB controller to reset the power to the zwave stick
Tony Tam
@fehguy
OK i'm on 1.4.3031
I would think that if it were in the device, a restart of the app wouldn't help. I would need to restart the device with a power off or special command
this is an Aeotec gen 5 fwiw
Elias Karakoulakis
@ekarak
ok - I'll try an older ozw version from http://old.openzwave.com/downloads/
Tony Tam
@fehguy
for me, it takes about a day before hitting this state
Elias Karakoulakis
@ekarak
I've fallen into the black hole that is a Node.JS upgrade
Elias Karakoulakis
@ekarak
right - I've rebuilt my local instance using Node 10.1 and OpenZWave 1.5
Tony Tam
@fehguy
ok cool, @ekarak, I'm also trying out reducing the polling to see if I'm just overwhelming my network
Tony Tam
@fehguy
OK @ekarak i've disabled polling on all my nodes so we'll see if the issue is resolved. I ran into an annoying issue where my dimmers will report their value right after it begins to change, so the value reported isn't the final value. To handle that, I'm starting to poll on value change, and stop after 15 seconds. Other than that, nothing is polled now so the network should not be congested at all
Elias Karakoulakis
@ekarak
okay - lets see how we fare in the following days
Tony Tam
@fehguy
thanks for the help
Elias Karakoulakis
@ekarak
np
Tony Tam
@fehguy
ok @ekarak, there is some really weird stuff happening when polling is disabled. The system gets a push notification when switches are toggled but only for a short amount of time. That means, after a few minutes, changes to the device state are not detected by OZW. If I set the poll interval to be really long, they always report their state. So I'm not sure what is up, but I can't disable polling or device state is just lost. Oddly enough, if I trigger an update of the device through OZW, the device state changes are detected for a few minutes. Then it just ignores them
Nothing has stopped working since I made this change last night, so we'll see if reducing the polling interval solves my initial issue. If not, I'll add some logic to disconnect from OZW and reconnect when it gets stuck. Feels better than restarting the whole system, which lives in a docker container.
Tony Tam
@fehguy
Hi @ekarak , just a follow-up on this. Reducing the polling interval helped the system run longer, but eventually I ended up with the same state--some nodes would eventually be unreachable until a reboot of the software. I ended up adding a crontab to reboot nightly--if you have any other ideas, let meknow
Chuong Le
@chuongle
hi everyone, I just start working with open zwave and I hope you guys can help me with a couple questions:
  1. Is it possible to delete a user code from a zwave lock? I can't find a way to delete it. Some mention you can pass 0000 as the code to delete but it only works with certain locks. I'm testing on a Schlage Connect 469 and Kwikset 916. None of the work. Is this because of how the device is built or is this because the OpenZwave does not support it?
  2. I follow the example in the repo and the scan complete fire multiple times until it finishes scanning all the devices in the mesh network. Is there an event to detect when whenever it finishes scanning all devices?
Daniel Lando
@robertsLando
The problem you faced has been resolved in 1.5.6, in the screenshot you sent here you was building 1.5.5 that was bugged
About your last error, which version of Openzwave are you using? 1.6 or 1.4?
I suggest you to use openzwave 1.4 that is more stable, openzwave-shared is still not ready for Openzwave 1.6
vibr77
@vibr77
Hi All, is there a problem using node worker with openzwave , I got an exception when running in a thread worker ?
vibr77
@vibr77
It seems that openzwave shared is not supporting Node worker thread with NAN... it has to be ported to N-API... am i right.
vibr77
@vibr77
? need to use cluster instead of worker thread...