Problems With Aeon Labs Z-Stick on VMware, Fails Most Times [Solved]

[HomerSimpson]

Hi,

Firstly, thanks to the developers for the huge amount of work that has obviously been put into this project. The more you start to scratch the surface the more you can appreciate just how much has been done.

Unfortunately though, I've got a problem I'm struggling with and need to ask for help. My situation is that I'm currently setting up a very small Z-Wave network (one Z-Stick, and one light dimmer) as a proof of concept before I go nuts and automate the whole house - I'm not prepared to take the plunge without knowing whatever technology I use will be rock-solid, and I'm afraid I'm not there yet. The issue I have is that most times I start the LMCE core, the Z-Stick fails to initialise, and it simply doesn't work. It seems quite random though, and on rare occasions it will start correctly and work perfectly, same hardware, same software (more often than not it fails though). I'm running a 10.04 snapshot on a VMware virtual machine, and this is a sample of the log output:

== ATTEMPT FRESH START ==
1   02/24/13 20:01:16   /usr/pluto/bin/Spawn_Device.sh 53 (spawning-device)   11606 Dev: 53; Already Running list: 15,16,18,19,29,30,
== FRESH START ==
1   02/24/13 20:01:16   /usr/pluto/bin/Spawn_Device.sh 53 (spawning-device)   device: 53 ip: localhost cmd_line: ZWave
0   02/24/13 20:01:16   53 (spawning-device)   Entering 53
========== NEW LOG SECTION ==========
1   02/24/13 20:01:16   53 (spawning-device)   Starting... 1
1   02/24/13 20:01:16   53 (spawning-device)   Found /usr/pluto/bin/ZWave
01   02/24/13 20:01:16.635      TranslateSerialUSB pci0000:00/0000:00:11.0/0000:02:00.0+2.1 result /dev/ttyUSB0 <0xb773d6d0>
Return code: 255
3   02/24/13 20:01:18   53 (spawning-device)   Device died... count=1/50 dev=53
Sun Feb 24 20:01:18 EST 2013 died
========== NEW LOG SECTION ==========
1   02/24/13 20:01:29   53 (spawning-device)   Starting... 2
1   02/24/13 20:01:29   53 (spawning-device)   Found /usr/pluto/bin/ZWave
01   02/24/13 20:01:29.289      TranslateSerialUSB pci0000:00/0000:00:11.0/0000:02:00.0+2.1 result /dev/ttyUSB0 <0xb77926d0>
Return code: 255
3   02/24/13 20:01:30   53 (spawning-device)   Device died... count=2/50 dev=53
Sun Feb 24 20:01:30 EST 2013 died


...and it just keeps repeating until it hits 50 attempts and gives up.

Any suggestions would be appreciated!

[iberium]

its generally a vmware issue handling the usb.  i run into this issue on other system when i have to plug in to program them.  im not sure exactly how this happens but sometimes vmware will not remount the usb under the same id, even sometimes telling you that it is connected but really showing up as a null device.  i would suggest that in a normal install this wont be an issue, as running your core in a vm is going to create headaches that you wont have if it is on a dedicated machine.

[Armor Gnome]

Compared to other commonly available automation interfaces/comm methods.  Z-Wave for me has been absolutely stable, reliable and easy to set up.  I have however from time to time had similar issues with the Aeon-Labs Z-Stick acting strangely on a new install.  So much so once I almost tossed it out thinking I damaged it.  Below are some quick suggestions for that specific device.

Do not connect the USB stick until you reach the Sara Wizard. 
Do not connect the USB stick to any USB hubs,  even some USB 2.0 PCI cards have given me issues.
My best success with that piece has been using older USB 1.0 ports.

For inclusion of future devices I suggest getting a cheap Zwave controller.  Mine is battery powered and runs about $5-$10 on eBay.  To pair new devices I can leave the USB stick in my core and pair the new device in its permanent location.  After transferring the device information to my Z-Stick, I remove the controller from the devices list and put the batteries back in a recharger.

Finally, take this into consideration if you wish.  There is nothing wrong with 'testing something' under VMs.  Personally I love checking out the latest at distro watch and seeing what's new.  For automation though and something as complex as LMCE, it truly does make everything smoother and ultimately more satisfying to use dedicated hardware.  IMO.

[HomerSimpson]

Thanks for the feedback. The plan is to run the 'production' version on dedicated hardware anyway, so for the moment I won't get too worried about it. Indeed with a bit of unplugging and re-plugging of the stick, bumping the virtual USB controller back to USB 1 and fiddling around along similar lines, it seems possible to get it to come alive okay, and once started correctly it works fine.

I'll report back if there's any continuing problems once it's running on hardware..!

[phenigma]

I have had good success with usb devices under virtualbox, as long as their data rates are not too high.

J.

[HomerSimpson]

Thanks again for the replies. FWIW, I was looking for an excuse to install and try VirtualBox, and phenigma's suggestion finally got me over the line. Indications so far is that it works without a problem under VirtualBox, so if anyone's considering running / trialling it on virtual hardware, perhaps go for VirtualBox over VMware.

[phenigma]

I'm glad to hear it is working for you!

Jason.

[hari]

I had a lot of issues with vbox and z-wave stick pass-through