Home Assistant Z-Wave Network Viewer (using Docker)

bar · November 1, 2020, 5:32am

Please let me know if I can add something on the original docker to make it work as before

squirtbrnr · November 1, 2020, 1:40pm

@bar something’s not right with my graph. It shows a node as dead and disconnected, but in HA it is not and I can control it just fine. in the graph it says there are no neighbors, but in HA there are 13 neighbors including the controller.

I have tried deleting all of the files the docker creates when it starts and starting a new docker container, but that did not fix anything.

geoffkrie · November 1, 2020, 5:09pm

I’m still unable to get this working. I just get a big black screen

INFO:Managers.data_manager:Connecting to ws://192.168.0.187:8123
ERROR:Managers.data_manager:Failed to connect due to error: [SSL: WRONG_VERSION_NUMBER] wrong version number (_ssl.c:1123) [LN: 87]

sjthespian · November 1, 2020, 8:01pm

Does your SSL certificate have the IP address set up as one of the valid names? If not, I believe the viewer is fail the SSL validation when it attempts to connect. If you can, use the hostname instead and that should fix that error.

geoffkrie · November 1, 2020, 8:20pm

Im not using SSL so i changed that and tried again. Again , big black screen with veiwer title on top left. here is entire log.

[15:15:56] INFO: Starting HA Z-Wave Network Viewer on port 6123
INFO:Managers.configuration_manager:Configuration is valid, data: {'home_assistant_url': 'http://homeassistant.local:8123', 'home_assistant_token': 'eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzI1NiJ9.eyJpc3MiOiJmMDdlN2Y3ZWU4OTU0OTExYTY3Nzc4YzZkYWFlNjhiMyIsImlhdCI6MTYwNDI0Nzg2NiwiZXhwIjoxOTE5NjA3ODY2fQ.-L9YHTEFFnfjynOEhn3rpGdC_SYn6iXv-tg-X5K-h1I', 'ssl_key': '', 'ssl_certificate': '', 'is_debug': False, 'is_local': False, 'server_port': 6123, 'home_assistant_web_socket_url': 'ws://homeassistant.local:8123', 'is_ssl': False, 'ssl_context': None}
INFO:Managers.data_manager:Connecting to ws://homeassistant.local:8123
 * Serving Flask app "webserver" (lazy loading)
 * Environment: production
   WARNING: This is a development server. Do not use it in a production deployment.
   Use a production WSGI server instead.
 * Debug mode: off
INFO:werkzeug: * Running on http://0.0.0.0:6123/ (Press CTRL+C to quit)
ERROR:Managers.data_manager:Failed to connect due to error: [Errno -2] Name does not resolve [LN: 87]
INFO:werkzeug:192.168.0.170 - - [01/Nov/2020 15:15:59] "GET / HTTP/1.1" 200 -
INFO:werkzeug:192.168.0.170 - - [01/Nov/2020 15:15:59] "GET /data/nodes.json HTTP/1.1" 200 -
INFO:werkzeug:192.168.0.170 - - [01/Nov/2020 15:15:59] "GET /favicon.ico HTTP/1.1" 404 -
INFO:Managers.data_manager:Connecting to ws://homeassistant.local:8123
ERROR:Managers.data_manager:Failed to connect due to error: [Errno -2] Name does not resolve [LN: 87]
INFO:werkzeug:192.168.0.170 - - [01/Nov/2020 15:16:54] "GET / HTTP/1.1" 200 -
INFO:werkzeug:192.168.0.170 - - [01/Nov/2020 15:16:54] "GET /data/nodes.json HTTP/1.1" 200 -
INFO:Managers.data_manager:Connecting to ws://homeassistant.local:8123
ERROR:Managers.data_manager:Failed to connect due to error: [Errno -2] Name does not resolve [LN: 87]

sjthespian · November 1, 2020, 8:24pm

Sorry, I saw the SSl error and didn’t see that the URL the viewer was logging was just ws://xx. @bar
any ideas on that error?

From the other error, your system doesn’t know the host name so you will have to use the IP.

How are you trying to connect to the viewer? Make sure you have just http in that url as well and not https unless you have enabled ssl in the configuration.

geoffkrie · November 1, 2020, 10:11pm

I have tried both ways. Here is the error with the ip address.

NFO:Managers.data_manager:Connecting to ws://192.168.0.187:8123
ERROR:Managers.data_manager:Failed to connect due to error: [SSL: WRONG_VERSION_NUMBER] wrong version number (_ssl.c:1123) [LN: 87]

bar · November 2, 2020, 5:53am

Please PM me the files so I’ll be able to reproduce it at my env.

Thanks

vloris · November 4, 2020, 10:45am

I’m trying this for the first time and it seems I have the same ‘SSL: WRONG_VERSION_NUMBER’ error message. I’m running home-assistant core in a docker container, and ha-zwave-network as well.

Both homeassistant and ha-zwave-network give error messages, presumably because ha-zwave-network is trying to use ssl, and homeassistant because it isn’t.

ha-zwave-network    | INFO:Managers.configuration_manager:Configuration is valid, data: {'home_assistant_url': 'http://homeassistant:8123', 'home_assistant_token': 'secret', 'ssl_key': '', 'ssl_certificate': '', 'is_debug': False, 'is_local': False, 'server_port': 6123, 'home_assistant_web_socket_url': 'ws://homeassistant:8123', 'is_ssl': False, 'ssl_context': None}
ha-zwave-network    | INFO:Managers.data_manager:Connecting to ws://homeassistant:8123
ha-zwave-network    |  * Serving Flask app "webserver" (lazy loading)
ha-zwave-network    |  * Environment: production
ha-zwave-network    |    WARNING: This is a development server. Do not use it in a production deployment.
ha-zwave-network    |    Use a production WSGI server instead.
ha-zwave-network    |  * Debug mode: off
ha-zwave-network    | INFO:werkzeug: * Running on http://0.0.0.0:6123/ (Press CTRL+C to quit)
ha-zwave-network    | INFO:Managers.data_manager:Connecting to ws://homeassistant:8123
ha-zwave-network    | ERROR:Managers.data_manager:Failed to connect due to error: [SSL: WRONG_VERSION_NUMBER] wrong version number (_ssl.c:1122) [LN: 87]
homeassistant       | 2020-11-04 11:37:20 ERROR (MainThread) [aiohttp.server] Error handling request
homeassistant       | Traceback (most recent call last):
homeassistant       |   File "/usr/local/lib/python3.8/site-packages/aiohttp/web_protocol.py", line 275, in data_received
homeassistant       |     messages, upgraded, tail = self._request_parser.feed_data(data)
homeassistant       |   File "aiohttp/_http_parser.pyx", line 523, in aiohttp._http_parser.HttpParser.feed_data
homeassistant       | aiohttp.http_exceptions.BadStatusLine: invalid HTTP method

So, it seems the two containers can talk to each other, I just have to convince ha-zwave-network to use plain html.

–edit–
It looks like the ssl options in the configuration_manager are not being honoured when the connection is setup here and it always tries to connect using ssl…

johntdyer · November 6, 2020, 1:09pm

yea, I am seeing the same thing… Seems there is no way to turn it off

bar · November 6, 2020, 1:30pm

I will fix it later this weekend, thanks for the details

bar · November 6, 2020, 3:10pm

Released new version with a fix, can you pls check?
Thanks

geoffkrie · November 6, 2020, 5:27pm

I uninstalled and then installed from the addons and it is working perfectly. Thank you.

geoffkrie · November 6, 2020, 5:31pm

After checking the log i see this warning:

WARNING:Managers.data_manager:Invalid device 922dadbf9b2c47aaaac15356c47d8d9e identifiers: []

how do i track down what device that is?

bar · November 6, 2020, 6:16pm

Look for that id in .storage/core.device_registry
Please update if that’s a valid device

Thanks

squirtbrnr · November 6, 2020, 8:35pm

@bar, I figured out my issue of the random node. the device registry had an old entry for a node that I had removed, then re-added to the zwave network at a later date. This is why it shows as node 9 on the map, but in HA it’s actually node 45. for some reason, HA did not clean up the device and remove it.

sjthespian · November 6, 2020, 8:53pm

FWIW, you don’t need to uninstall and reinstall it if you’re using the addon, you can just hit the “rebuild” button and it should pull down the latest version from github.

geoffkrie · November 6, 2020, 11:08pm

I had tried so many combinations of configurations including a rebuild a few days ago so i got frustrated and uninstalled. When i saw that Bar had fixed what looked like my problem , I re-installed. it’s all OK now.

mirekmal · November 7, 2020, 8:26am

Few weeks ago I removed the component, since after one of upgrades it stopped to display anything (json was empty). Today I reinstalled the latest version and it display my network properly… but also something more. So for the background; since last installation I made significant changes in my zwave network configuration, that forced me to remove and re-add most of nodes. I made cleanup of zwcfg_xxx.xml and also, using Zen Tools, I cleaned all dead nodes on directly on the stick. All past ghost devices are no longer showing in entities, integration configurator or in zwave viewer (one without python version). To my surprise however, they popped once again in this add-on as failed nodes!

On the screenshot above all these disconnected nodes exist in my network, but with different node numbers and names (after reconfiguration) and only this addon is able to display these old ghosts… so the question; is this information cached somewhere, so it can survive several HA restarts and even addon reinstallation? Any way to clear this? I have no clue, after all the cleanups I made, where information about these nodes could pop up from…

bar · November 7, 2020, 10:56am

can you please PM me a link to the debug files (how to available in README),
I’ll reproduce it in my testing environment and update you with the finding,
if you would like to try first, please remove the docker container (in case you are working with hassio - uninstall the addon), make sure the relevant debug volume was removed as well, and then install it again,
An alterntative solution can be to access that directory in the container and delete the files and restart the container.

If it still happen, it means HA still contain those devices (as ghost).

Please let me know if you need more assistance