In the meantime I just wanted to clarify a couple of items. Because you have a keyboard that includes a usb 1.1 hub and you are now testing with two different usb 2.0 hubs it is not perfectly clear in some of your descriptions above when you say hub which one you mean so I'll ask you the following:
1. Does the first USB 2.0 hub, the Maxxter device with card reader, work correctly with Ultibo when the USB Keyboard/Hub combo is NOT connected?
2. Does the second USB 2.0 hub, the Vivanco device, work correctly with Ultibo when the USB Keyboard/Hub combo is NOT connected?
3. Do I understand correctly that the USB Keyboard/Hub combo works briefly sometimes when connected to the second USB 2.0 hub and then stops working but never works when connected to the first USB 2.0 hub?
I'm sure you are being very methodical about your tests, I just need to make sure I am following it correctly from afar so we don't waste time with misinterpretations.
mvdhoning wrote:Doing this with the keyboard attached trough the usb hub only makes a partial dump and then freezes halfway the dump. So no txt file for that one.
The usb hub however does work under linux. Somehow the usb makes the usb part in ultibo freeze when trying to access it?
Regular readers will know that whenever someone says freeze or hang we always want to know more information about the manifestation of that freeze or hang. In most cases a single thread hanging while performing a certain function (like the USB dump to file) will be due to either a deadlock (waiting to acquire a lock that another thread is not releasing) or an exception (such as accessing a null or invalid pointer). In either case it is likely that the rest of the Ultibo core is still happily running and you can use that to assist with diagnosis.
In the case of an unhandled exception then a log entry will often (although not always) be generated to alert that the thread has been halted due to an exception, you can find references on how to enable logging in the Forum Quick Reference.
A thread that is simply waiting to acquire a lock that will never become available can be harder to see but using the WebStatus unit (also found in the Forum Quick Reference) allows looking at the state of each thread and in particular the one you asked to perform the USB dump.
None of the above will provide an instant answer to the problem, they just help to narrow it down further.
mvdhoning wrote:With this one unplugging and replugging the usb keyboard helps a bit as sometimes i get a working keyboard for about 5-10 seconds and then it is 'dead' again.
In this case it is possible that the keyboard is going into an error state and being disconnected after 5-10 seconds which is why it stops working, any USB error should also be reported in the logging output as described above.