Device is not hot plugged / patched to Universe

[dandaka]

1. I have an OLA client that always sends data to OLA (Afterglow library). It is running.
2. Launch OLA
3. OLA doesn't see USB-DMX device (Eurolite USB-DMX)
4. Unplug and plug back USB device
5. Still no results
6. When I open OLA web UI
   • I see this device, but it is not connected to Universe 1 http://take.ms/GPHRe
   • Expected: The device gets back to set Universe, when I plug and uplug the device
7. Patch the device to the Universe, all is working

Log https://gist.github.com/dandaka/bd2bf0974ac9f54a9d42

Another log, where I had only to replug USB device without going to Web UI
https://gist.github.com/dandaka/c441b706fc986cce25ae

[nomis52]

Likely a dup of #125 .

[peternewman]

Were you able to confirm @dandaka ? Was this a duplicate?

[dandaka]

@peternewman Don't have an access to hardware anymore.

[peternewman]

Okay thanks @dandaka .

Looking at the log @nomis52 this actually looks more subtle.

https://gist.github.com/dandaka/bd2bf0974ac9f54a9d42#file-gistfile1-txt-L546
olad/plugin_api/DeviceManager.cpp:105: Installed device: EurolitePro USB Device:12-eurolite-20-24
olad/plugin_api/PortManager.cpp:119: Patched 12-eurolite-20-24-O-0 to universe 1

https://gist.github.com/dandaka/bd2bf0974ac9f54a9d42#file-gistfile1-txt-L904
olad/plugin_api/DeviceManager.cpp:105: Installed device: EurolitePro USB Device:12-eurolite-20-25

https://gist.github.com/dandaka/bd2bf0974ac9f54a9d42#file-gistfile1-txt-L1315
olad/plugin_api/DeviceManager.cpp:105: Installed device: EurolitePro USB Device:12-eurolite-20-26

It appears to be specific to the Eurolite, or other devices with no serial and the way we assign a name to them as bus and device address:
https://github.com/OpenLightingProject/ola/blob/master/plugins/usbdmx/EuroliteProFactory.cpp#L67

See also:
#1149

In comparison with other similar devices we only support one device (as they don't have a serial), but it always patches to the same universe. https://github.com/OpenLightingProject/ola/blob/master/plugins/usbdmx/DMXCreator512BasicFactory.cpp#L52

It's a balance between how many devices with no serial we can support at once, versus how reliably they appears in the same place. Having one per bus may be a reasonable compromise, assuming most machines have a few USB hubs.

[peternewman]

If you replugged every time to fix the device @dandaka then that likely explains the new device address.