Mine was fine for quite a while after the initial reports until a reboot. Then I had to apply the patch and was ok, then it stopped again for while, then started working again.
MyQ is definitely making some changes. Theory is it may not be consistent across all servers.
Silver lining is the pymyq library was picked up by @Lash-L and getting some overdue attention.
At this point I have my all local solution finally implemented, so it doesn’t matter too much.
My x86-based HA running on Proxmox was updated to Z-Wave Server 1.32.1 and Driver 12.0.3 last night and everything is back to normal after 2 days without Z-Wave working.
A lesson learned, I suppose. I always wait until the next point release before installing a new version of HA but I have always gone ahead and installed newer drivers as they appear…
What I do on docker is
a) only upgrade 1 component at a time
b) shutdown ha and zwave containers
c) duplicate ha or zwave config directories (I put the version # in here)
d) update docker-compose with the new version to deploy and point to new Config directories
d) start up and see how it works.
e) if it doesn’t I go back to the prior versions and config directory.
f) if it seems to work I let it run for 48 hours before upgrading the next component.
Typically I update zwave first, then HA second.
I would think in Proxmox you could snapshot or duplicate the VM and instantly go back to that?
Thank you! Same issue here- running Home Assistant Core in a VENV and within an EXSI VM… like the others, I’m also curious to learn what the Soft Reset switch does and why this resolved the issue for us?
The soft reset is a command that basically is the equivalent of yanking power to the USB stick and turning it back on. In most instances that’s not an issue.
In the case of some hypervisors the USB port is recycled and because when using the Virtual USB port the port I’d isn’t guaranteed…
The net effect is the port vanishes and no more usb stick. Poof.
To prevent this when using one of the affected sticks (aeotec 5/5+ and many other 500 sticks) turn OFF soft reset AND specify the USB path as By-Id
Hi All, I have this same issue on Home Assistant 2023.10.1 running on unraid VM. Passed Aeotec Gen5 stick through as ACMO - but after some time the zwavejs add on stops with the error
2023-10-11T18:18:49.769Z CNTRLR No response from controller after 1/3 attempts. Scheduling next try in 100 ms.
2023-10-11T18:18:50.870Z CNTRLR Failed to execute controller command after 2/3 attempts. Scheduling next try i
n 1100 ms.
2023-10-11T18:18:52.973Z CNTRLR The controller is unresponsive
2023-10-11T18:18:52.973Z DRIVER Attempting to recover unresponsive controller...
2023-10-11T18:18:52.990Z CNTRLR The controller does not support soft reset or the soft reset feature has been
disabled with a config option or the ZWAVEJS_DISABLE_SOFT_RESET environment va
riable.
2023-10-11T18:18:52.991Z DRIVER Recovering unresponsive controller failed. Restarting the driver...
Error in driver ZWaveError: Recovering unresponsive controller failed. Restarting the driver... (ZW0100)
at Driver.destroyWithMessage (/usr/src/node_modules/zwave-js/src/lib/driver/Driver.ts:2790:17)
at fail (/usr/src/node_modules/zwave-js/src/lib/driver/Driver.ts:3505:14)
at /usr/src/node_modules/zwave-js/src/lib/driver/Driver.ts:3554:5 {
code: 100,
context: undefined,
transactionSource: undefined
}
Shutting down
Closing server...
2023-10-11T18:18:52.996Z CNTRLR [Node 009] ping failed: Timeout while waiting for an ACK from the controller (
ZW0200)
2023-10-11T18:18:52.997Z CNTRLR » [Node 009] querying node info...
2023-10-11T18:18:52.997Z CNTRLR » [Node 009] Assigning SUC return route...
2023-10-11T18:18:52.998Z CNTRLR » [Node 009] Deleting SUC return route...
2023-10-11T18:18:52.998Z CNTRLR [Node 009] Deleting SUC return route failed: The driver is not ready or has be
en destroyed (ZW0103)
2023-10-11T18:18:52.998Z CNTRLR [Node 009] Assigning SUC return route failed: The driver is not ready or has b
een destroyed (ZW0103)
Server closed
[18:18:53] WARNING: Halt add-on
s6-rc: info: service legacy-services: stopping
s6-rc: info: service legacy-services successfully stopped
s6-rc: info: service legacy-cont-init: stopping
s6-rc: info: service legacy-cont-init successfully stopped
s6-rc: info: service fix-attrs: stopping
s6-rc: info: service fix-attrs successfully stopped
s6-rc: info: service s6rc-oneshot-runner: stopping
s6-rc: info: service s6rc-oneshot-runner successfully stopped
I can manually disable the soft-reset setting - but it is not remembered after the addon is restarted. Does anyone know how do I get it to properly remember the Disabled setting? I tried the UI and yaml (through the UI) methods but it doesn’t work…and I am not sure how to access the zwavejs config directly (the Addons SMB share is totally empty )
Yes I’m pretty sure I’m hitting the right save button - it’s normally greyed out, and I see when I change it to Disabled the Save button at the bottom of that section turns blue and is clickable and I get the message that it needs to restart zwavejs. But it always reverts to Automatic. Is there some way to directly edit the zwavejs config file though ssh etc?
It could also be that this is not the problem… it always seems to work until it tries to talk to Node 029, maybe it’s somehow borked and crashing zwavejs or the stick. But not sure how to really troubleshoot this.
Hello, I think I faced the same problem. However I had a warning about my power supply in the logs, and so I really thought this problem was mince, not a software issue. I have replaced my power supply, and now… all my zwave devices are gone.
I tried to revert to previous version of zwave, of core etc… to no avail (also disabled the auto reset as suggested above)
I am back to the latest releases. I can see my gen5 stick, but it is the only device available.
Did I lose it all in that mess, do I need to re-pair everything, or is there a way to recover my zwave network? Thanks!
I’ve been having somewhat slow responses with zwave from day 1. Whether it’s not enough repeaters/routers i’m not sure. It’s been solid just sometimes there’s a 1-2 second delay. All the gear is aeotec multisensor6 and z-stick gen5.
When you updated your z-stick do you remember what version is was?
I’m currently running zwave js (not started) and zwave JS UI as a front end only with mqtt disabled.
The version of the zwave stick is:
AEON Labs Z‐Stick Gen5 USB Controller ZW090
FW: v1.1
SDK: v6.51.10
I’m curious if this the same version as yours when you updated? Did you have to update through windows and did you notice any improvement/change after updating?
I’ve currently toggled the soft reset option off as others have above to get around the addon(s) not working.
It’s occasions like this that confirm my approach regarding addons such as zwavejs. “Don’t update unless the update is fixing a problem you are experiencing”. The same applies to HA generally.
I do agree & I was always very careful with my upgrades and made sure I have a snapshot to roll back. At the same time, I have to admit that for last ~1-2 years Zwave upgrades were smooth.
Hello @reedy,
I do not recall what version I was running.
I think you can only upgrade to the windows, so yes sadly I used windows to upgrade. It started working right away after upgrade. Through the troubleshooting I disabled soft reset and it stayed that way.
Apologies for delayed response, I was not home over the weekend.
My aeotec Z-Stick Gen5 ZW090 is still running firmware 1.1, and does not work with zwave-js staring with version 12.0.0, released on September 26, 2023.
I don’t have a windows machine available to upgrade firmware at the moment, so I spent some time downgrading zwavejs to see where it broke.
Downgrading to 11.4.3 restores full functionality. This should probably be a bug report on the zwave-js github, but I’m starting here.