Zwavejs2mqtt: [question] z-wave network and range extenders

[bushvin]

I recently moved all my devices from a defective z-wave controller to a new one (gen 7 silabs SLUSB700A UZB-7).
I added 2 range extenders, but I have the impression they are not working, or not relaying the signal where it needs to go.
Unfortunately for me, the signal test is sent back to the controller as a Power Level Report, which I have the impression is not (yet) supported by zwavejs. So I can't tell if coverage is adequate.

When I move my Raspi zwave controller to the area with problems, all devices report as normal. Should I leave my controller where it is so all devices can try to figure out their way? Does z-wave heal itself and re-establish routes when it is able to connect to the controller? Will I loose control if I move my controller back to the place I want it to be, as it was before I moved it to this area?

Is there another way for me to test the range extenders?
When interviewing and healing them, there are no errors, or at least there don't seem to be any errors in the log.

Where would I start troubleshooting this?

[robertsLando]

Did you try to run a network healing? That should update return routes on every device. I dunno if you should also setup some associations in order to make extender to work @AlCalzone may help with this

[AlCalzone]

The latest version (4.2.0, driver 7.4.0) has an improved healing strategy and also makes sure that the network map is up to date. I'd start by looking at a log of the healing process in that version.

[bushvin]

upgraded to 4.2.0
I seem to be affected by this bug
I had to re-interview all the battery-powered nodes and manually wake them up (to speed up the process)

When launching Heal network all battery-powered devices were reset to asleep and never active.
I'll let it run for a while, waking up all nodes manually to see what happens...

[andyboeh]

I have an Aeotec Range Extender 6 and it works flawlessly without any configuration (on a 500 series controller). My device has a LED that blinks whenever data is being relayed, so it's easy to tell if it's working. I cannot remember whether it worked before healing the network.

[bushvin]

After fixing the earlier mentioned bug (thanks again, @robertsLando and @AlCalzone ), I was able to proceed:

I looked for a spot where I expected to have the best coverage everywhere, and proceeded to perform a network heal. I had to wait until my battery-powered devices woke up to finish the heal, so this might have gone faster if I first had reset the sleep time to 5 minutes or so. Waking them up manually would yield varying results, and having all devices take care of it automagically worked best, apparently. Z-Wave is a bit like Novell networks: have patience, give it a chance to fix itself.

So with a successful heal, I proceeded to move the controller to a spot closer to where it would end up (roughly halfway). I reasoned that since all my nodes already had multiple paths, they would be able to reconnect and improve any routing using updated neighbors.
I performed another network heal, and again I waited for a little over an hour for it to finish normally. Result: success! All nodes could connect to the controller. (And vice versa)

Next step: put the controller in the intended location followed by another network heal and another hour of waiting.
Lo and behold: I have a fully functional z-wave network with each node having multiple routes to the controller!

I don't know if it would have worked if I omitted the 2nd heal, it might have, if all routes somehow ended up with the Controller.

Anyway, thank you all for chipping in! I hope this may help anyone at some point.

[robertsLando]

Thanks for your feedback glad it works now! I think the second heal was needed, everytime the controller is moved the routes should be updated to find the best ones