Can I use results from ping or polling with Z-Wave JS to see if the network is functioning?

Thanks for categorizing people who try to help you finding a year long problem you had as noise ! Makes one feel really good about helping you out ! Have a nice day.

Actually you didn’t try to help me at all with the question I asked, you started right off the bat asking why I wanted to do this and explaining why you personally wouldn’t do it (even using that exact word). So I tried to be patient and I spent a bit more time than I wanted to explain to you why I wanted/needed to do it this way. I have been doing automation for about 30 years, and my solutions are rarely by the book, but they work in real life, both in my house, my cabin, my cars, my boat and on my motorcycle. As long as the end users (me, my family and the people who rent 2/3 of my house) never notices anything going wrong, that’s what I call working in real life, no matter what band-aids I am using to achieve it.

Because this is a perfect example of the XY problem. You have to understand from our shoes, we have no idea what your skill level is. Unfortunately, this means most people who help (including myself) always assume everyone a basic user using HassOS. Turns out that you’re not a basic user and you’re using core. How would we know that off the bat without asking questions that you deem below yourself? Lastly, as a person who’s been an engineering role w/ software development for 15 years and worked with this home automation engine for over 7 years, the best solution is to fix the problem instead of hiding it. Otherwise you’ll be back here again with the same question when something else changes and this band aide no longer works.

Either way, the only way to actually set up polling in Zwave JS is to use ZwaveJS2MQTT and enable the polling ability through MQTT. If you’re just running the ZwaveJS server on your os, you’ll have to look up the commands to poll a device and do that after startup. There’s no way to do it from the HA integration yet, it’s all set in ZwaveJS, which can only be done through ZwaveJS2MQTT currently.

1 Like

Right, this forum software doesn’t show how long you’ve been on a forum (2018) or how man posts you have (it’s too busy dishing out those stupid badges). So it does make it harder to see that I’m not a newbie at first sight. Alex was right off the bat passive-aggressive “Personally it would drive me insane to have a system that locks up every few weeks / months, but OK that’s me.” And then the rather condescending tone continued with “What I don’t get is why do you need to constantly ping a zwave switch (and generate totally unneeded network traffic on an already low bandwidth mesh) to check if your Pi/HA crashed ?” Maybe that’s what annoyed me? There was no attempts at all to actually answer the question I asked, which I thought was pretty clear, “how do I ping something on the Z-Wave network”.

This las answer of yours, on the other side is perfect! That tells me exactly what I need to know, and I will look into that this weekend, since it probably demands changing some setups in ZwaveJS2MQTT, which I won’t do unless I’m able to concentrate fully on what’s going on, in case something happens. Work takes too much time until that. So thanks a lot for that tip! :+1:

I’ve never used the polling ability. I’m pretty sure you just enable it on the node in question, then you make an automation in home assistant and use the service described in the link. It may not work that way, it may behave differently. You’ll have to figure that out through trial and error.

Actually I will probably poll directly from the keep alive system in EventGhost instead, less back and forth. :grin:

That wasn’t even necessary to do! Turned out that polling the device type (it polls all devices of the same type when it polls, which confused me a bit - I found that out on GitHub) was enough for the old system to work (I had automations that sendt feedback from the thermostats to a log in EventGhost and was a part of the keep-alive system too, only they were to infrequent to be reliable for a pulse signal). So I’ll probably set it to two minutes since when polling the different thermostats uses variying time to answer, so all in all that should cover about every minute.