My purely debian-family Linux Chromebook has been acting weird lately. I suspect that maybe when I replaced the internal battery a while ago, some of the connections inside might not have been put back properly because it was very difficult to reassemble. I was very careful but totally possible I made an error because I'm not too experienced with that. All the little ribbons and clips were so fiddly.

Is there a way to monitor for loose connections from the terminal? Some kind of log entries that would be clues? I just am looking for the basic info here, don't feel anything is too obvious!

Problems I'm having are like pauses of unresponsiveness, and a lot of confusion about the power (battery vs ac, full vs empty, plugged vs unplugged). Bluetooth has become less reliable also. Generalized poltergeist type behavior.

Edit: I have confirmed a hardware problem. Pressing the keyboard and the bottom plate together around the power-in helps. So something is loose in there.

all 9 comments

sorted by: hot top controversial new old
[–] 6 points 1 day ago (2 children)

You could take a look at the kernel log by running the dmesg. This might give you some clues if some drivers are encountering issues.

As another commenter said, bad battery seems likely for the charge level symptoms. For the bluetooth, it might be the antenna that got disconnected or severed.

Easiest to check by opening it up again. But if you're afraid of doing more damage you could take it to a confident friend or some repair shop.

  • source
  • hideshow 2 child comments
  • [–] [S] 1 point 1 day ago (1 child)

    So I would be looking for messages about drivers?

    Would that indicate a hardware problem or that I have the wrong drivers? I don't understand what you mean.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 6 points 1 day ago (1 child)

    Technically, anything is possible, but I think you'll encounter more catastrophic failures if you have loose connections.

    Why not just disassemble and double check the connections? Ruling it out early will save time while debugging the software side.

  • source
  • hideshow 1 child comment
  • [–] [S] 3 points 1 day ago

    I thought everything was good when I out it together. I don't know how to do a better job. It was very difficult to put back together so I could just make it worse with another attempt. If it is possible to do some reconnaissance in advance then I will prefer that.

  • source
  • parent
  • [–] 4 points 1 day ago*

    Is there a way to monitor for loose connections from the terminal?

    The systemd journal does the monitoring for you.

    Just enter journalctl -b for this boot, locate relevant time spans. Maybe read man journalctl to figure out how to find what you're looking for.

    There's also dmesg which shows you the kernel's ring buffer. But I find most if not all entries in the journal as well and if I'm not mistaken "ring buffer" means that new entries overwrite old entries.

    If you can reproduce the problem, start journalctl -f or dmesg -w first, then reproduce. That should be helpful.

  • source
  • [–] 4 points 1 day ago

    Man, my whole life seems like loose internal connections.

  • source
  • [–] 4 points 23 hours ago

    'acting weird' is not enough of information. Try to write down everything that is happening without guessing the cause at first.

  • source
  • [–] 2 points 21 hours ago

    Technically no. No software could ever detect a bad connection unless the spec has a sense method for every data line before communicating. And even then, it would still be inconclusive. But the result of bad/intermittent connection is indeed usually quite severe, the sign can be negligible at first tho. Off spec timing, bit flip, occasional garbled data, all of which could stall or halt the system in weird ways. It doesn't hurt to do a reassembly.

  • source