515. Very briefly: boot issues and nvram (anecdote)

I had to move offices at work last Friday (which partly explains my radio silence).

One of my nodes, with a Biostar 880G+ motherboard, wouldn't boot. The issue manifested itself in a few different ways (and not always at the same time):

Most of the time the bios portion of the boot would be very slow, and I'd get a loooooong beep when the voltages were echoed during what I presume is POST. I never made it to the GRUB screen. I'd had this happen once or twice before, especially when the computer had been turned off and/or unplugged for a longer period of time.

Occasionally I'd get a message about NVRAM being checked..

After
1) having checked that nothing had vibrated loose during the transport
2) having removed all cards and drives except a single RAM stick
3) having checked that RAM stick
I still wasn't getting any closer to getting booted. So I went into the BIOS. Having seen the message about nvram and having done a bit of reading online I selected to erase the NVRAM.

And suddenly booting worked...once.

After that I kept getting messages about nvram being corrupted and boot fail, or 'NVRAM...Update failed'.

I then set the CMOS jumpers to pin 2 and 3 (small jumper next to the CMOS battery)

And then it worked! It booted just fine! But I kept getting the following error echoing on my screen:

[  120.718683] hub 3-0:1.0: unable to enumerate USB device on port 2
[  120.932378] usb 3-2: new full-speed USB device number 28 using ohci_hcd
[  121.072497] usb 3-2: device descriptor read/64, error -62
[  121.316673] usb 3-2: device descriptor read/64, error -62
[  121.556904] usb 3-2: new full-speed USB device number 29 using ohci_hcd
[  121.697019] usb 3-2: device descriptor read/64, error -62
[  121.941224] usb 3-2: device descriptor read/64, error -62
[  122.181425] usb 3-2: new full-speed USB device number 30 using ohci_hcd

Unplugged all the USB devices. Turned out to be the KVM (ritmo). Turning it off (i.e. unplugging all USB input to the KVM so that it lost power) and plugging it in again solved it.

And ever since then all has been fine.

Presumably it's time to change the CMOS battery.