USB Zwave Stick crashing

[Brononius]

For one or the other reason, my USB Zwave GEN5 stick fails after the update of proxmox.
This device is used for an openHAB Virtual server.
It worked well for over a year, but since a few days/week (probably after the latest update), it suddenly stop working. I don't see anything in the logs that could explain this. So no disconnection or so on the proxmox server, or on the openhab server. Sometimes it happens after a hour, sometimes after a day. So not clear why/how.
I've installed a physical server with openHAB and the stick, and there it works fine. So I guess it's an issue with the hardware or with the proxmox. Personally I guess proxmox since it happens after the updates.

Any idea what I can do to troubleshoot/resolve this?

What I've done in the past:
Check out the exact ports:

lsusb -t
/: Bus 06.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/2p, 5000M
/: Bus 05.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/2p, 480M
/: Bus 04.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/6p, 5000M
/: Bus 03.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/14p, 480M
|__ Port 1: Dev 2, If 0, Class=Human Interface Device, Driver=usbhid, 1.5M
|__ Port 1: Dev 2, If 1, Class=Human Interface Device, Driver=usbhid, 1.5M
|__ Port 2: Dev 3, If 0, Class=Vendor Specific Class, Driver=gspca_zc3xx, 12M
|__ Port 4: Dev 5, If 0, Class=Hub, Driver=hub/4p, 480M
|__ Port 1: Dev 6, If 0, Class=Communications, Driver=cdc_acm, 12M
|__ Port 1: Dev 6, If 1, Class=CDC Data, Driver=cdc_acm, 12M
|__ Port 9: Dev 4, If 1, Class=CDC Data, Driver=cdc_acm, 12M
|__ Port 9: Dev 4, If 0, Class=Communications, Driver=cdc_acm, 12M
/: Bus 02.Port 1: Dev 1, Class=root_hub, Driver=ehci-pci/2p, 480M
|__ Port 1: Dev 2, If 0, Class=Hub, Driver=hub/8p, 480M
/: Bus 01.Port 1: Dev 1, Class=root_hub, Driver=ehci-pci/2p, 480M
|__ Port 1: Dev 2, If 0, Class=Hub, Driver=hub/6p, 480M

Assign the ports to my virtual machine:

qm set 1123 -usb3 host=3-9
qm set 1123 -usb4 host=3-4.1

Afterwards, on the VM machine, they're used in the application openhab with:

crwxrwxrwx 1 root dialout 166, 0 Oct 26 08:43 /dev/ttyACM0
crwxrwxrwx 1 root dialout 166, 1 Oct 26 08:41 /dev/ttyACM1

 

[Brononius]

Anybody an idea what I can do?
It's been 3 weeks now that I'm trying to troubleshoot/solve this issue.

I've installed serveral OS (redhat, centos & mint), and each time the same. After a while, suddenly the usb stick doesn't work anymore. Nothing in the logs.
I've replace the disk with Proxmox and installed a standalone centos, and the problem didn't take place. It ran for about 6 days without any hassle. I restored the orginal config (CentOS under proxmox and latest updates). Sadly the problem is back after about 3 hours. So it's really something to do with Proxmox on this machine.

Maybe something with that usb driver?
If I analyze it correctly, then proxmox sees a xhci_hcd devices, and the VM a cdc_acm?
Not sure if I can force both ways the same? Even not sure if it's a good idea.


All suggestions are welcome!

 

[fireon]

Please use SSD/HDD for Proxmox install. USB is only a for playground. And... usb is single point of failure, yes you can have two of them and Raid... but USBsticks are never a good solution for a OS that was not desinged to run on it. You can search the forum here. There are a lot of threads about some similar problems.

 

[Brononius]

The device(s) aren't a hard disk.
It are 2 AEON GEN5 Zwave sticks. Call it a kind of wireless dongle to connect zwave devices to your server.
So I really need this to connect a part of the home automation to a VM machine (openhab2). :$

 

[fireon]

The device(s) aren't a hard disk.
It are 2 AEON GEN5 Zwave sticks. Call it a kind of wireless dongle to connect zwave devices to your server.
So I really need this to connect a part of the home automation to a VM machine (openhab2). :$

Oh ok sorry ;( i understand you complettly wrong.

xhci is USB3.0. This doesn't really work on 5.x. Tested last week with an normal USBstick. If you connect the sticks to an 2.0 port, it should work. Can you test this please?

 

[Brononius]

The nodes itself are USB1 it seems (12MB). And the port is an USB2 (480MB). Not directly anything on USB3/5000MB, except the driver? At least, that's my conclusion...

From lsusb:

lsusb -t
/: Bus 03.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/14p, 480M
|__ Port 1: Dev 6, If 0, Class=Communications, Driver=cdc_acm, 12M
|__ Port 1: Dev 6, If 1, Class=CDC Data, Driver=cdc_acm, 12M
|__ Port 9: Dev 4, If 1, Class=CDC Data, Driver=cdc_acm, 12M
|__ Port 9: Dev 4, If 0, Class=Communications, Driver=cdc_acm, 12M

For reference?

12M = 12MBit/s = USB1
480M = 480MBit/s = USB2
5000M = 5000MBit/s = USB3

So the driver (xhci) is normally for USB3, but in my case on a USB2 port?

 

[fireon]

Please delete the USBentry from your config and add it via gui. Test it if it is working. You have an checkbox to enable/disable usb3.0. Diff the entry that is set in the config. Here are one cut from us vm's.

agent: 1
bootdisk: scsi0
cores: 4
cpuunits: 1023
description: Proxmox PC
hotplug: disk,network,usb,memory
memory: 4096
name: kassa.tux.local
net0: virtio=51:CF:39:D3:24:35,bridge=vmbr0
numa: 1
onboot: 1
ostype: win10
protection: 1
sata0: cdrom,media=cdrom
sata1: none,media=cdrom
scsi0: local-lvm:vm-200-disk-1,discard=on,size=32G
scsihw: virtio-scsi-pci
smbios1: uuid=88ba20bb-b4da-3938-b2dd-2a3e43b121aa
sockets: 1
usb0: host=04b8:0202
usb1: host=0c2e:0a01
usb2: host=08e6:3438
usb3: host=1-13,usb3=1
vga: qxl

 

[Brononius]

Gave it a try, but both option (usb3=1 or usb3=0) ended up the same: crash after a while... :'(

It's very strange, since it worked fine for about a year, and a while ago (a month or so), the problems started. Guess it start happening after an update... I rarely update, and now I know again why I don't like to update... :$

 

[fireon]

Hmm, maybe it is a kernelthing. There are updates on your system? When yes what updates?

 

[Brononius]

The updates have been done through proxmox.
So not sure what has been updated. Only the timing is a bit the same when the issues started.

 

[Brononius]

Do you think that this issue could be resolved in the nearby future?
Today, I've got 2 hardware servers running in order to let the HomeAutomation work all time. Not optimal for power consumption.
So or I need a solution in/with proxmox, or I need to migrate to something else.

 

[fireon]

I think the only way is to stay on your working pveversion. Fill also out an Bugreport!

 

[mcflym]

Hi there,

is it possible that it's connected to this issue I mentioned?:

https://forum.proxmox.com/threads/p...ct-frame-counter-not-updating-disabled.38074/

 

[RobFantini]

The updates have been done through proxmox.
So not sure what has been updated. Only the timing is a bit the same when the issues started.

perhaps check apt logs at /var/log/apt/history.log* , see if there was a kernel upgrade at the time the issue started, if so try prior kernel.

 

[aC23]

Sounds like a similar issue I'm having: https://forum.proxmox.com/threads/dev-ttyacm0-not-showing.60977/
Were you able to get it working on a USB 2.0 port?