1

My Raspberry Pi 2 B (rev a01041) has been working great for ages. I usually have two hard drives plugged into it. Yesterday, I plugged in another one. Today I plugged in a fourth. Immediately, the Pi started behaving oddly in my ssh session, spitting out errors as below.

$ vim ~/.zshrc
zsh: Input/output error: vim
prompt_vcs:53: vcs_info: function definition file not found                                             
$ sudo reboot
zsh: Input/output error: sudo
prompt_vcs:53: vcs_info: function definition file not found

I couldn't ssh in another session. I then pulled and replaced the power plug to hard boot it. It wouldn't boot. The power and ACT LEDs are solidly on, but there are no other signs of life. No ethernet LED, no HDMI signal, etc. I tried removing the two extra hard drives but that didn't make a difference. Finally, I replaced the Pi with another B+ (rev 0010), and it booted up fine.

  1. Is my Pi permanently dead?
  2. Could plugging in too many USB devices have killed it (is there no protection)?

EDIT: further troubleshooting

As per suggestions from Steve Robillard in the comments, I re-tried booting the broken Raspberry Pi 2 B with no hard drives connected at all. This did not help.

On a whim, I plugged in the spare SD card that was previously in the B+ (with OSMC on it). Oddly enough, this time I did get the four-colour-pixel startup screen, although it froze there. The red power LED lit up, but the green ACT did not.

Community
  • 1
Sparhawk
  • 693
  • 4
  • 18
  • 33
  • Have you tried removing all of the drives? IU assume that all of the drives were connected directly to the Pi, not a powered USB hub. – Steve Robillard May 19 '16 at 12:43
  • Drawing too much current may trip the main polyfuse at the microUSB port -- it certainly has affected people here before. Leave it unplugged until tomorrow and then try again. – goldilocks May 19 '16 at 12:49
  • @SteveRobillard I didn't previously, because there's a bit of fstab and config things involved. However, I tried it as suggested and re-edited. – Sparhawk May 19 '16 at 12:55
  • @goldilocks I thought that the polyfuse would prevent everything working, but I suppose it depends on how much it's tripped? My second question is because I've just placed an order for a new Pi 3, and I'm wondering if there is a danger in connecting the same four hard drives. Also, whether the 2.5 A power supply with mitigate that risk. – Sparhawk May 19 '16 at 12:58
  • 1
    Sorry, I was going on the title and stopped reading at "...It wouldn't boot." Bad habit. Only halfway through my first coffee. ;) I think you are correct; the other reports involving the polyfuse said no lights at all. – goldilocks May 19 '16 at 13:27
  • 1
    I would still use a powered USB hub. It is cheap insurance and avoids under powering the drives. – Steve Robillard May 19 '16 at 13:44
  • @SteveRobillard I was concerned about lowering overall throughput when all drives are running. Also, all drives are 3.5" self-powered; would that mean that they don't actually draw power from the Pi? – Sparhawk May 19 '16 at 22:09
  • 1
    I don't see how a USB hub hurts throughput. Do the drives have a seperaate plug or wall wart? – Steve Robillard May 19 '16 at 22:11
  • @SteveRobillard If the USB2 connection is the limiting factor for throughput, then connecting four of them through one (Pi) port would force them to share this bottleneck. Yes, each drive has a separate power supply. – Sparhawk May 19 '16 at 22:13
  • 2
    how is that any differnt than plugging them all into the pi's ports which is absically a hub – Steve Robillard May 19 '16 at 22:14
  • Without looking at the internals of the drives or a schematic I can't say if they pull power over the USB cable or not. – Steve Robillard May 19 '16 at 22:15
  • @SteveRobillard Ah, I was not familiar enough with the internals of the Pi. I thought that there were four separate USB controllers. Thanks for that information. – Sparhawk May 19 '16 at 23:19
  • 2
    Note the USB bus is also shared with the Ethernet port so that is likely to be your bottleneck. – Steve Robillard May 19 '16 at 23:29
  • Oh! @goldilocks, I tested it again today, and it's fine. Does that confirm it must have been the polyfuse then? I'm a scientist by profession, so plugging the four hard drives back in is tempting. If it wasn't permanent damage the first time, should it not be permanent the second time? – Sparhawk May 21 '16 at 11:54
  • I've never set one off. However, I think one of the reports here that turned out to be a polyfuse thing involved similar symptoms in that initially it did not work at all, but then a short time later the lights came one but it did not work -- implying the polyfuse may let through some current when it has not reset completely or has not been tripped sufficiently hard. But I'm not familiar with the board schematics or the nature of polyfuses enough to say whether that's true or not, and by "I think" I mean that could be a garbled/false memory... – goldilocks May 21 '16 at 12:13
  • ...A lot of the polyfuse discussion tends to go on in comments like this and they aren't searchable. So if it is working today, the polyfuse seems the most likely explanation, and it would be good (since no one else has left an answer) if you left one to that effect (since answers are searchable, then this can be easily found again via "polyfuse"). In 24 hours you'll be able to tick it as accepted. – goldilocks May 21 '16 at 12:15
  • @goldilocks I'm more than happy to do this. I've just edited the title to make it more searchable. However, if you would like post a minimal answer, I'm happy to accept and upvote. – Sparhawk May 22 '16 at 02:06
  • Actually re-reading this I'm not totally confident about that; it was something I threw out based on symptoms, but I would have thought it impossible to do via USB. My understanding of DC power regulation is rudimentary at best, so I've asked this question to see if we can get some more information. – goldilocks May 22 '16 at 18:02
  • Thanks @goldilocks, it also seems odd to me. I've caved to my curiosity and I'll try plugging in the four hard drives again when I can spare the downtime… for science. – Sparhawk May 22 '16 at 23:17

0 Answers0