No… Regarding the zwave specific, I agree there is something different. Since 2.5 (first snapshot) It has always chosen ACM0 for some reason.
But it is mainly controled by the Linux behind. So if you have other devices using the same ACM0, you´ll risc that they change port in OH.
By creating a symlink, you´ll be able to tell through OH (PaperUI) exactly what device to be use, wether or not Linux had exchanged their physical port adresse.
Thanks Kim. I have nothing else connected to RPi4B than the Z-Stick and the LAN cable so OH should suggest the correct port right away, and I feel very strongly about that. Mind you, if this does not work then the very first step in adding a z-wave device (the very controller itself) to RPi4B running OH2.5.0 fails. Not very user friendly at all, and should be fixed!! So again, how can we remove this critical obstacle? Do you know how it can be brought to the attention of the right people?
True, the PaperUI initiates the detection. Semantics.
I agree with you. Unfortunatly thats not the way Linux works. So I dont think there is any easy options here.
It may be possible to create somekind of an utility which scan the ports and letting the user decide, and then this utility writes the symlinks automatic. But I doubt anyone is going to do this.
Hi guys,
had the same issue and my problem was the serial binding (before 2.5 it seemed to work, then I had to rename the binding). I am still not sure if the serial binding (serial1) is still needed in order to get Zwave binding working with an usb-stick. Please ignore my post if it’s not the case…
Regards,
Herbert
OK. So your feeling is that the OH community will accept that OH2.5.0 cannot run on a RPi4B when the commonly used Z-Stick is the controller. Hmmm, scary. Thanks for the discussion. Take care.
BTW: I get numerous RPi3B+ OH 2.5 warnings saying:
2020-01-10 13:40:43.476 [WARN ] [essage.ApplicationUpdateMessageClass] - TODO: Implement Application Update Request Handling of New ID Assigned (64).
Do you happen to know what it means?
The community happens to find a solutions to make it work.
My apologies!
You were right. It’s the Stick that causes the problem with RPi4. After a few more hours of searching I came across this github post: https://github.com/raspberrypi/linux/issues/3027, and the first entry on Sept. 20, 2019 explains why. There is even a test report attached from Aeotec on the issue. Interesting reading if you are a hardware guy. I’ll go get me a cheap USB2 hub and see if it works. Thanks.
1 Like
Did the original poster get his problem solved?
The issue with the Aeotech stick and Rpi4 is not the same as the original poster, cause he uses an Rpi3.
Solved the issue at my site. 
1 Like
Just one question to get it right:
The issue is with RPi4 and the Aotec zwave stick because of USB2/3 incompatibility on the stick’s HW.
So it’s (most likely) NOT related to low (USB) power?
Because I recently (after switching to OH3 on RPi4 - previously 2.5 on RPi3) experienced slow zwave reponse on the very stick above and use an SSD instead of a sd-card, I thought insufficient power might cause this as well!?
The older Aeotec sticks appear as USB3 devices but do not work with USB3. Using a USB2 hub inline is a workaround instead of replacing the stick with a newer one.
If you use a powered hub that would ease the power supply load too,
Any indication what means „older“?
My stick is from 2016, but seems to work quite well on the usb2 Connector of the RPi4.
I recognized some lag while switching nodes, but in general it’s ok.
I am wondering if if works differently on the USB3 connector!?
IF the stick is working well serially then I do not think a hub would help.
Does your zwave network have any zombie nodes in the network that do not actually exist? Thank can mess up zwave routing causing delays. One other issue would be devices close to the limit of the zwave signal.
I recently removed Zombie nodes with the external (windows) tool - don’t recall the name.
SiLabs PC Controller. I think it is now part of the Simplicity Studio free download.
1 Like
Right - that’s it 
1 Like
Hi NCO,
Just wanted to say that the Aeotec has admitted there is a Z-Wave stick issue and has released an updated version named Z-Stick Gen5**+**. I have not been able to get my hands on it yet but would love to get rid of my USB dongle.
But you probably knew this already.
Anyway, Take care and enjoy the summer!!
1 Like
Thank you very much
For letting me know.
I will check whether I should get one.
Have a great summer, too 
I’ve actually had similar issues, but in my case it’s not Pi related. I have an x86 server running Ubuntu 20.04 and OH runs in a docker container. I’ve had z-wave issues for awhile, so long that I’m not sure when exactly they started (it was sometime late 2.5 or early 3.0 for me).
My stick fails to connect to openHAB, usually after a restart of the computer or container. I’ve noticed some conflict with my Monoprice whole house amp which uses the Serial binding. I have discovered that switching off the amp, restarting, then powering the amp usually lets everything connect. Sometimes it takes several restarts and cache clearings though.
I also discovered that the stick does not work with USB 3 ports.
It’s certainly an irritating problem, but one that I haven’t posted about simply because I do manage to get it reconnected. Perhaps I’ll look into a new bridge device if Aeotech has admitted the hardware is causing problems.
@satman , please let us know if the updated stick works better for you!