USB hangs Windows 2000 guest vb6 program - 2 updates

[PrairieMutt]

Running host Linux 3.16.0, Debian 8.5 Jessie, XFCE 4.10 64 bit (4096 MiB) and guest Windows 2000 pro sp4 32 bit (2498 MiB) with guest additions. Virtualbox 5.1.4 r110228 Qt 5.3.2.

Application in question is a vb6 sp6 program. It ran fine for many months. Then recently, it just crashed. It is possible that it was due to a security fix for USB drivers for Debian. After trying many things, including new hardware, reboots, guest additions reload, etc., I was unable to get the guest to ever recognize the USB connection to my NI USB-6008 data acquisition module. Aside from keyboard, mouse, there is a only the USB connection to the NI USB-6008 data acquisition module using NI driver 2.0.0f0, the latest that will run on Windows 2000. Upgraded from 5.0.24 to 5.0.26 and USB and all seemed to work OK. But the vb6 program will now hang after two to 12 hours. The vb6 program simply becomes inactive and will not respond to user input. It consumes no CPU. Other applications on the guest work fine, but after interacting with the guest, sometimes the whole (host) XFCE becomes unresponsive and I have to open a root terminal (ctl-alt-F1) and reboot.

So I upgraded to 5.1.4. Same problem. I tracked it down to some strange interaction with the USB. If I manually (for now, instead of using vboxmanage programmatically) disconnect the NI USB connection from the guest via the USB icon at the lower right of the guest VM window, the program then responds normally (normally for having lost its USB connection), responds to user input, complains (correctly) that it is not getting data, etc. Then I reconnect the USB to the guest at the lower right of the VM windows, tell my program to go look for the USB-6008 again, it finds it and goes its way working perfectly until the next random hang. I reiterate that this was not a problem with 5.0.24 which ran without either host or guest reboot for months. I searched google and these fora in vain. There is nothing about this in the manual. I would appreciate whatever help anyone might give me to fix this without the kludge of removing and then restoring the USB connection from host to guest. Very many thanks.

In the log file attached, you can see where I did the detach, then reattach twice. Note that there is nothing in the log that shows any abnormalities prior to that time ... so far as I can see, which might not be very well. Windows 2000 Event log reveals nothing. Thanks again.

UPDATE: I restarted the program in question in the IDE. The next time it hung, I entered the break hot-key sequence. I then disconnected the USB from the guest as above. The IDE then came alive and broke on a command using the API for the NI USB. Whether the guest-host USB interface breaks, then cause a hang at the next access or whether the USB access randomly causes the guest-host USB interface to break, I have no idea. In any event, the log shows nothing in this case as well. Thanks again all.

UPDATE 9/2/16: I am also getting occasional Blue Screens of Death (BSOD) on openhci.sys, which is a Microsoft USB driver. I checked IRQs and driver status info in /control panel/system/device manager and all appears OK. Using IRQ 9. BSOD details: Stop: 0x000000D1 (0x08F4EC08, 0x00000002, 0x00000000, 0xF630B45C) Address 0xF630B45C, base at F6308000, date stamp 3e5ffecb - openhci.sys.

There is definitely some kind of USB problem that has been introduced at some point. Many thanks.