For a little while now, my thinkpad randomly has the following issue while running Debian 10.8 and I have no useful resources online as to how to resolve it. I am running it on a Thinkpad T480

enter image description here

  • What's the question? The sda device seems to be dying. Or it is underpowered and misbehaves. Or data connection cable is broken. See SMART at least, and please reformulate your post as a question. Feb 13, 2021 at 18:41
  • Sorry, we can't accept images of code or errors. Post those as text, so that we can try to reproduce the problem without having to re-type everything, and your question can be properly indexed or read by screen readers.
    – DavidPostill
    Feb 13, 2021 at 18:44
  • @DavidPostill I feel this is too strict in this case. It is possible they can't just copy it as a text, this picture seems to be a phone photo. But, to help properly we need to see exact dmesg messages. It is quite cruel to force users to retype it. Feb 13, 2021 at 18:51
  • @NikitaKipriyanov It's meant as a suggestion ...
    – DavidPostill
    Feb 13, 2021 at 19:01
  • @DavidPostill if I can get directions or pointers on how to get my system to print those logs to a file so that I can paste it here, I would be more than happy to.
    – modernNeo
    Mar 12, 2021 at 6:56

1 Answer 1


Usually a message about an I/O error means there is a problem with the drive. While it is possible that problem is a firmware problem, cable problem, or a problem with power, usually it means the drive is dying.

You can boot off a live CD and run smartctl -t short /dev/sda (package smartmontools if it's not already installed) and then, once the test is done, run smartctl -l selftest to see the results. Usually this will provide you with an overall summary of the state of the drive, but if not, you can run smartctl -x to see that.

In my case, it says this:

SMART overall-health self-assessment test result: PASSED

which means my drive is in a good state. I suspect yours will not say that, and you will need to replace your drive. There is no way to effectively repair a broken drive without specialized equipment, so replacement is it.

If the drive reports that it's good, you could then look into things like changing out the cable or querying the power supply via /proc to see if those are the cause of the problem. But, in my experience, those are less common than drive mortality.

  • thanks, I will do that first and see what the results are.
    – modernNeo
    Feb 22, 2021 at 17:56
  • I tried that and I also got PASSED. do you have any suggestion on what to look at next?
    – modernNeo
    Feb 22, 2021 at 20:11
  • Then it could very well be the cable to your drive. If that's bad, it can definitely cause failures.
    – bk2204
    Feb 23, 2021 at 1:55

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.