Closed canedje closed 3 years ago
Same here:
Message: Z-Wave JS: Setup erneut versuchen
USB Device is unknown:
Thank you in advance
I had exactly the same error
The bad part is that even the VM was corrupted not recognizing the USB device!!
I'm not able doing a rebuild like you did. A restore via ha backup didn't work, you said too.
Hope the version 1.47 can repair this issue...
canedje @.***> schrieb am Fr., 5. Nov. 2021, 21:44:
[image: image] https://user-images.githubusercontent.com/23150947/140576150-69508c06-dc42-458a-8563-158e4b1fd2e9.png I had exactly the same error
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/home-assistant/addons/issues/2252#issuecomment-962208822, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR4KG2Z3PZRTEOZZT3T4LYTUKQ63NANCNFSM5HOUVEXQ . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.
I only did get it running to reboot my NUC (Running Unraid with dockers and VM) first going back to a running full restore of HA(not the partly zwave-js snapshot)
Great hint. I'll try to restore a my last full backup and reboot my ECM Server with unRAID too. Thx
canedje @.***> schrieb am Fr., 5. Nov. 2021, 21:54:
I only did get it running to reboot my NUC (Running Unraid with dockers and VM) first going back to an running full restore (not the partly zwave-js snapshot)
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/home-assistant/addons/issues/2252#issuecomment-962213838, or unsubscribe https://github.com/notifications/unsubscribe-auth/AR4KG24NC6U24NBPKRRKI7DUKQ77RANCNFSM5HOUVEXQ . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.
Same issue here. Back Up of .45 and reboot fixes the issue, but impossible to upgrade to .46.
Now it's working again. But stupid of me. I reset the Zwave Stick earlier. now all my 64 devices and over 700 entities are gone ... I thought these are stored in the home assistant.
All my automations are dead. I might throw up. Damned.
Never touch a running system. Old saying, but wise.
0.1.46 upgraded the z-wave driver. The new version will perform a "soft-reset" of the z-wave controller upon startup. This isn't compatible with all systems, because it can result in a removal and re-insertion of the USB device. The soft-reset can be disabled, but the addon does not provide that option yet.
Now it's working again. But stupid of me. I reset the Zwave Stick earlier. now all my 64 devices and over 700 entities are gone ... I thought these are stored in the home assistant.
All my automations are dead. I might throw up. Damned.
Never touch a running system. Old saying, but wise.
That’s my worst nightmare
0.1.46 upgraded the z-wave driver. The new version will perform a "soft-reset" of the z-wave controller upon startup. This isn't compatible with all systems, because it can result in a removal and re-insertion of the USB device. The soft-reset can be disabled, but the addon does not provide that option yet.
Causing a lot of work and stress
I experience the same; after the soft-reset I can see the USB stick being disconnected from the virtual machine (Synology VM). If I reconnect the USB to the VM Zware-JS starts the init again and soft-resets the USB again resulting in a disconnect. Restored 0.1.45 and no issues.
I experience the same; after the soft-reset I can see the USB stick being disconnected from the virtual machine (Synology VM). If I reconnect the USB to the VM Zware-JS starts the init again and soft-resets the USB again resulting in a disconnect. Restored 0.1.45 and no issues.
Exactly the same issue :(
Yeah, this just screwed me over too. Same as others report - also attempting to restore from backups (thankful for such a good integrated backup system). But from what @kpine says this shouldn't be too complex a hotfix to resolve?
Following
A few days ago I noticed the exact same behaviour. I'm running HassOS as a VM on Unraid and the USB device was constantly being disconnected. The weird part is that my Decon container running directly on Unraid (and NOT in HA) also stopped working with an error that the USB device (ConBee II) was not reachable. After several reboots of both Unraid, Docker and HassOS VM, I got it to work (both for Deconz and Z-Wave JS). I think it was the timing between mounting the USB device in Unraid and restart of the Z-Wave JS to MQTT add-on that did the trick.
Unraid version 6.9.2 Home Assistant OS version 6.6 Home Assistant Core version 2021.10.7 Z-Wave JS to MQTT version 0.28.0
If you can switch to the community add-on it offers an option to disable the soft-reset.
What do you mean "switch to the community add-on"? I do use the standard zwave-js add-on in supervisor of HA? I do not see this option
@canedje The community add-on repository has an alternative zwavejs add-on (zwavejs to mqtt) with a few extra features and the option to disable soft reset . The mqtt is optional.
I have the same problem. Removing and adding the stick works for some time and after that it crashed again. I restored my backup of 1.45 for now... Hopefully this will be fixed in the 1.47 version where the soft-reset can be disabled.
@cogneato I tried that before, but did not get it up and running. Thanks for the answer I hope there will be a solution for the normal Zwave-JS addon
This also affects me, so glad I had a snapshot to revert to.
The "Soft Reset" option found in ZWaveJS2MQTT definitely needs to be exposed in the official ZWaveJS add-on configuration.
Same issue here, I reverted back to 1.45 for now. I really hope this will be fixed as soon as possible.
I have turned off Soft Reset and I think I can confirm it's working. I have restarted Z-Wave JS to MQTT, as well as detach and attach the USB stick in Unraid, and I don't see any unexpected "Hardware Removed" in the logs. For now, it's working like a charm!
Not sure how and if this could have affect Deconz as well. Perhaps it's just one hell of a coincident, or something happens with the USB controller in Unraid.
Same here! Had a lot of trouble. Had to rollback to .45 and restart NUC. The Z wave stick was glowing red after the update.
When I´m using "zwave js" version 1.45 AND Z-wave JS to MQTT in Version 0.28.0 > same error. (Based on core-2021.10.6) < Core 2021.10.7 and above has an issue with "SolarEdege API"... No fun these days
zwave js 1.45 and Z-wave JS to MQTT Version 0.27.0 works (Based on core 2021.10.6)
BUT
zwave js 1.46 and Z-wave JS to MQTT Version 0.27.0 WORKS!!!
The problem isn´t zwave js 1.46 - it is the Version 0.28.0 from Z-wave JS to MQTT.
When I´m using "zwave js" version 1.45 AND Z-wave JS to MQTT in Version 0.28.0 > same error. (Based on core-2021.10.6) < Core 2021.10.7 and above has an issue with "SolarEdege API"... No fun these days
zwave js 1.45 and Z-wave JS to MQTT Version 0.27.0 works (Based on core 2021.10.6)
BUT
zwave js 1.46 and Z-wave JS to MQTT Version 0.27.0 WORKS!!!
The problem isn´t zwave js 1.46 - it is the Version 0.28.0 from Z-wave JS to MQTT.
No, it's not because I don't use the "Z-Wave JS to MQTT" add-on at all and it was broken since I upgraded from "Z-Wave JS 1.45" to "Z-Wave JS 1.46" and it works again since I rolled back to version 1.45.
I have the same issue. Rolling back to version 1.45 made it work again. Can someone implement an option to disable the "soft reset" of the Z-wave stick?
Same issue and same fix (rolling back to 1.45).
Please refrain from posting "same issue" and similar replies without adding new information. Every new post sends a notification to all people subscribed to the issue. The appropriate way of showing that you're affected by the issue is by reacting with 👍 to the top post.
Thanks!
If you're running your VM on libvirt (KVM) as is pretty common, you can use udev rules to automatically reattach the USB when it gets disconnected. I wrote a small guide here.
In theory, this is better that just disabling the soft-reset, as I assume there are good reasons for doing so or they wouldn't have added that "feature".
I am also on unRaid running HASSOS in a vm and downgrading to 1.45 and rebooting the vm didn't solve the issue.
@kekonn you might need to remove the USB passthrough and add it back.
I have. I also physically unplugged it, checked the vm, then added it back in. I switched to an external zwavejs2mqtt this morning as a workaround.
I have. I also physically unplugged it, checked the vm, then added it back in. I switched to an external zwavejs2mqtt this morning as a workaround.
I am also on unRaid running HASSOS in a vm and downgrading to 1.45 and rebooting the vm didn't solve the issue.
You need to restart unraid for it to work, just restarting the VM won't do the trick.
Strangely, when I downgrade (I can only downgrade to 1.4.0), I get immediately upgraded again automatically to 1.4.6. Auto update is turned off. So I don’t really have a workaround (using Synology virtual machine)
In theory, this is better that just disabling the soft-reset, as I assume there are good reasons for doing so or they wouldn't have added that "feature".
To be fair, ZWaveJS has implemented soft reset knowing that a) it breaks certain configurations, but more importantly b) it provides a mechanism to override/revert to the previous behaviour. The next version of ZWaveJS actually tries to auto detect problematic ZWave controllers to prevent the soft reset from occurring.
It’s the “official” HA ZWaveJS add-on that does not appear to have taken either point into account.
The next version of ZWaveJS actually tries to auto detect problematic ZWave controllers to prevent the soft reset from occurring.
This already exists in the version of node-zwave-js that that addon has upgraded to. In addition to detecting specific controllers that are known to be problematic, the driver is also supposed to remember if a soft-reset fails, and avoid it in future startups, but that isn't working as intended. If it was, at worst it would have been a one-time failure (with perhaps a host restart) and correctable w/o changing any settings. Unfortunately, there isn't the manual fallback to workaround it.
I've opened a PR that allows the user to disable the soft reset from within Home Assistant: https://github.com/home-assistant/addons/pull/2260
In https://github.com/home-assistant/addons/pull/2261 we've added detection of VM in use and automatic disable of soft-reset. We want to allow users to get running as easy as possible without configuring options as far as possible.
An option for advanced users, that can configure their VM and want to use soft-reset, is ok to add but it must be compatible with the automatic handling. :+1:
Please test version 0.1.47 when it shows up and report the success/failure. Thanks!
In #2261 we've added detection of VM in use and automatic disable of soft-reset. We want to allow users to get running as easy as possible without configuring options as far as possible.
An option for advanced users, that can configure their VM and want to use soft-reset, is ok to add but it must be compatible with the automatic handling. 👍
Please test version 0.1.47 when it shows up and report the success/failure. Thanks!
Works perfect. But had to restart HA server after update to .47. Thank you allot 🤝
0.1.47 works for me as well (Synology VMM). Much appreciate the fix.
Closing here since 0.1.47 seems to resolve the issue. We'll continue to try improve how soft-reset is handled and welcome help with that.
0.1.47 tested with hass 2021.11.2 on ESXi 6.7. Issue resolved. Thank you very much for the fix!
0.1.47 Tested with hass 2021.11.2 on ESXi 7.0U2a. Issue resolved. Thank you very much for the fix!
Yes 1.47 did the trick. Thanks!!!
Yup, same here. If anyone still needs an option to turn it on or off manually, let me know.
Hi guys, Maybe i'm not suppose to yet but i cant see the 0.1.47 update?
Try refreshing the add-on store.
Thank you so much guys,
I can confirm that 0.1.47 is working fine.
UnRaid 6.9.2 Hassos 6.6 VM
Same here, the update seems to have fixed the issue.
The problem
Zwave-js update to 1.46 did heavely crash the system. Even a restore of a backup is not working Updating the VM running HA gives an error the zwave USB is not recognized anymore. I had to run a working backup at HA and restart the NUC for getting up and running
Environment
System Health
Home Assistant Community Store
GitHub API | ok -- | -- Github API Calls Remaining | 4268 Installed Version | 1.16.0 Stage | running Available Repositories | 889 Installed Repositories | 29Home Assistant Cloud
logged_in | false -- | -- can_reach_cert_server | ok can_reach_cloud_auth | ok can_reach_cloud | failed to load: timeoutHome Assistant Supervisor
host_os | Home Assistant OS 6.6 -- | -- update_channel | stable supervisor_version | supervisor-2021.10.8 docker_version | 20.10.8 disk_total | 48.5 GB disk_used | 7.6 GB healthy | true supported | true board | ova supervisor_api | ok version_api | ok installed_addons | File editor (5.3.3), NGINX Home Assistant SSL proxy (3.0.2), Node-RED (10.0.1), Samba share (9.5.1), Terminal & SSH (9.2.1), Z-Wave JS (0.1.45), Samba Backup (5.0.0), ESPHome (2021.10.3), Duck DNS (1.14.0)Lovelace
dashboards | 3 -- | -- resources | 21 views | 97 mode | storageProblem-relevant configuration
Traceback/Error logs
Additional information