It wouldn't be an OPL problem, but USBHDFSD. It depends on the version that OPL was compiled with. OPL only tells USBHDFSD that it wants to write files, and that's it. Anyway, I think its problems are difficult to identify. We might need somebody with the domain knowledge of USB and a USB protocol analyzer, to understand and suggest fixes for the USB-related problems. I've got this feeling that there is some hardware-level compatiblity problem with some devices, since the errors are so random. But well, I've got no time or energy to bother with these things anymore. It is possible that when you don't do some things, the bulk of OPL can start up faster. Faster than the USB device can initialize, so much that the device is missed at boot. There isn't really a nice way to solve this because this is not a device that will be always immediately accessible after the related modules are loaded. In FMCB, I delay the startup process by 500ms, just to serve this purpose.