Disconnection GEN5 stick

Are there any others who experience disconnections with the ZWAVE GEN5 USB stick?
When I (re)boot the server, everything is working as expected. But after a while (sometimes 2 hours, sometimes 3 days), it stops working.

My setup:

  • Proxmox 4.10 as Virtualization Server
  • OpenHAB stable 2.1.0 on Virtual machine with CENTOS 7.4
  • 2 USB GEN5 sticks
    (one for my home, one for my garage (distance to far else, I noticed huge delays with 1 stick))

This week, I’ve bought a new stick because I thought it was maybe the stick itself. But after migrating all my house nodes (25), it worked for some hours, and then back the same result.
Since I use this USB stick with an USB converter over Ethernet (+/-60m), I thought this was the problem. But after putting the stick directly in the server, it keeps on failing after a while.

I was thinking of reinstalling my VM-server this week with fe an ubuntu server. But before I start this, I would like to know if you guys have another suggestion/remark/…

ps I can’t pinpoint the exact timing when the problem started. But a while ago, I reinstalled my complete setup to use the stable version of openHAB. Before I was using the nightly build version. And I’ve put in my mind that this could be a reason…

What hypervisor are you using? This is the behavior I was experiencing on ESXi 6.5 until I downgraded the hypervisor’s USB drivers. I’m on a GEN2 controller though.

I don’t think that would cause you problems with the USB over IP stuff though.

Proxmox (qm): https://www.proxmox.com/en/

I love this, because it has a complete admin GUI (http based) for easy management… :wink:

Grrr,

Last weekend, I’ve spend some time in reinstalling my whole server, and resetting the USB nodes. But no luck.
What I’ve done?

  1. Update Proxmox to latest version (5.0-34)
  2. Reinstall OH2 (stable version) on CentOS 7 as VM under Proxmox 5.0-34
  3. Reinstall OH2 (stable version) on Ubuntu 17.10 as VM under Proxmox 5.0-34
  4. Reinstall OH2 (unstable version) on Ubuntu 17.10 as VM under Proxmox 5.0-34

Some debugs on the step 4 today

ssh openhab@localhost -p8101
   password: habopen
   log:set DEBUG org.openhab.binding.zwave
   logout

tail -f /var/log/openhab2/* | egrp -i "zwave"

And it crashed/stopped after around 1 hour. Sadly I can’t see anything in the logs that could explain why.
If anybody could help me out with this riddle… :blush:

events.log: http://www.oniria.be/tijdelijk/events.txt
openhab.log: http://www.oniria.be/tijdelijk/openhab.txt

I’ll see today if I find the time/material to install a standalone server (not a VM) and check if this could solve my issue.

You don’t mention which version of the zwave binding you are using. I recall experiencing a similar crash after a few hours to a few days. IIRC, it would stop crashing if I did not load my Aeon HEMs. I don’t recall the build it was resolved in, but I do not get this with the development version of the binding. My setup is very different though (no VM, single controller, OH 2.2.0 snapshot).

I’ve installed it by default (over PaperUI).

via the karaf console:

bundle:list
196 | Active   |  80 | 2.1.0                  | ZWave Binding

When you got the crashes, did you see anything in your logs?

Did some digging and found an old post about it…

I had opened a ticket on Chris’ website and in the last post I stated this:

Since installing version 2.1.0.201706121815 of the binding, I have not had a crash. I think this may have been resolved, so I'll close the ticket.

You may want to try the development binding to see if that resolves your issue (scroll to the top of the thread I linked to).

Thanks!
I’ve just installed the version of last week.

cd /etc/openhab2/addons/
wget http://www.cd-jackson.com/downloads/openhab2/org.openhab.binding.zwave-2.2.0-SNAPSHOT.jar
ssh openhab@localhost -p8101
    feature:install openhab-transport-serial
    bundle:start org.openhab.binding.zwave
    bundle:list
       215 | Installed |  80 | 2.2.0.201710172006     | ZWave Binding

Will have a look how this run…

Didn’t same to resolve the issue. Instead, it also crashed other bindings (fe knx), but since i’ve done a lot of changes, hard to say if it’s really related with the snapshot version.

I reinstalled another physical server with ubuntu 17.10 and openHAB2 stable. And for the moment, it seems to be much stable. My zwave nodes aren’t properly in, but guess i’ll have to be a bit more patient for the wakeup cyclus.
So I guess that the problem will be related to the proxmox setup and/or hardware. tbc

This problem seems to be related with proxmox (debian). Driver issue with USB hardware.

I migrated about 4 weeks ago to VMWare, and since that time, no more problems…
So nothing to do with OpenHAB…

2 Likes