Is it possible to schedule a soft restart?
AFAIK there are no off-the-shelf means to achieve that.
Also, why would you want to do that?
Presumably, something keeps going wrong and a restart solves it.
I suppose you could rig up something with a shell script that you’d run once a day (or a few minutes after your integration goes unavailable), but the sustainable approach would be to find out and fix what’s wrong.
Off the top of my head, a short USB cord to get some distance between your HA system and the radio stick never hurts.
Use the home assistant supervisor: Restart add-on service in an automation.
I did this for a while when my sister had issues with zwave not working every few days. I just had it run in the early morning hours. At some point I turned it off and her issue had resolved.
Ooh, nice. I stand corrected.
But if you’re inclined to use that periodically, I’d still suggest to find the problem.
Any suggestions on how to make zwavejs more reliable and stop it from becoming unresponsive?
Trust me, I’d love for it to ‘just work’ as it causes nothing but frustration and has for a long time now.
For context I have 115 zwave devices. All of them have security disabled as a year or so ago I redid my whole network thinking that would help (never did obviously )
I do have a lot of zen15 that are very chatty by default. I’ll go through those and severely reduce the reporting to see if it has an impact.
Thanks this worked nicely.
For those looking for the automation, here is the YAML, please change the addon name as yours will have a unique string XXXXXXXX_zwavejs2mqtt
This restarts Zwave JS UI @ 3am nightly, which will reset the stick in the process.
alias: Restart Service Zwave JS UI Nightly
description: "Zwave JS UI Restart"
trigger:
- platform: time
at: "03:00:00"
condition: []
action:
- service: hassio.addon_restart
metadata: {}
data:
addon: XXXXXXXX_zwavejs2mqtt
mode: single
Would there be a way the automation can detect an unresponsive z-wave stick, so using this detection as a trigger?