If you see output similar to this, sometimes followed by the RasPi rebooting, this is often an indication of a power supply issue.
[ 13.141067] ERROR::dwc_otg_hcd_urb_enqueue:477: Not connected
[ 13.141067]
[ 13.197943] usb 1-1: USB disconnect, device number 2
[ 13.204725] usb 1-1.1: USB disconnect, device number 3
[ 13.211821] smsc95xx 1-1.1:1.0: eth0: unregister 'smsc95xx' usb-bcm2708_usb-1.1, smsc95xx
USB 2.0 Ethernet
[ 13.408078] Indeed it is in host mode hprt0 = 00021501
[ 13.658014] usb 1-1: new high-speed USB device number 4 using dwc_otg
[ 13.673842] Indeed it is in host mode hprt0 = 00001101
[ 13.938305] usb 1-1: New USB device found, idVendor=0424, idProduct=9512
[ 13.957988] usb 1-1: New USB device strings: Mfr=0, Product=0, SerialNumber=0
[ 13.988388] hub 1-1:1.0: USB hub found
[ 14.004766] hub 1-1:1.0: 3 ports detected
[ 14.308288] usb 1-1.1: new high-speed USB device number 5 using dwc_otg
[ 14.428542] usb 1-1.1: New USB device found, idVendor=0424, idProduct=ec00
[ 14.446881] usb 1-1.1: New USB device strings: Mfr=0, Product=0, SerialNumber=0
[ 14.481045] smsc95xx v1.0.4
[ 14.573406] smsc95xx 1-1.1:1.0: eth0: register 'smsc95xx' at usb-bcm2708_usb-1.1, smsc95xx
USB 2.0 Ethernet, b8:27:eb:bb:e1:62
The ethernet chip hangs off the USB bus is also the USB chip, so (although it does not show in the output of an 'lsusb' command) it is detected by the kernel the same time the bus is brought up, which causes an significant increase in the power requirements (40% as a guess but not actually measured it, someone with a nice bench power supply please do so).
After half a day of swapping power supplies, powered hubs, wiring in meters to monitor stuff etc. I found the problem, it was the USB lead itself that feeds the power into the RasPi causing a voltage drop. The moral is 'check the simple stuff that is easily tested first' it can save you hours.
EDIT: attempted over-simplification for clarity meant my post was misleading, so have re-written the offending sentence.
