On Sun, 9 Aug 2020 at 02:16, Tim via users <users@xxxxxxxxxxxxxxxxxxxxxxx> wrote:
Michael D. Setzer II via users wrote:
> So, no clue how the setting got changed??
While some may say a dying motherboard battery can be the cause for
CMOS or UEFI screw-ups, there are a plethora of possible causes.
Unless you can get it to happen again, you may never know. Just to go
through a few:
Even a good motherboard battery may have poor connection to the
terminals (dirt contamination, or poor tension against the metal). I
remember back in the 1980s being told to always wipe down coin
batteries before insertion, and not to get your finger grease on them
afterwards. While the latter is next to impossible without gloves, it
really just means don't contaminate the bits of metal on the battery
that touch the bits of metal in the battery holder. Though we were
also told that sufficient contamination across the batteries own
terminals would lead to it prematurely going flat, I think you'd have
to have particularly mucky hands to be able to do that.
Random circuitry behaviour when the mains voltage spiked or dipped (the
latter can cause some really odd things without causing damage). Logic
circuitry works by using high or low voltages to represent high (1)
bits or low (0) bits. It's either above or below a threshold. When
power goes down (a brownout) without fully turning off, circuitry can
end up with voltages in an level that's indeterminate. High mains
voltage spikes can outright destroy things, can stress and weaken
parts, or just induce voltages that interfere with operation at the
moment of the spike.
Random hardware faults that only rear their heads under a particular
set of conditions.
Software glitches that stomp over data in an area they shouldn't touch.
RAM faults.
The first step in understanding a problem is to focus on easy
mitigations and known hardware failure modes.
Visual inspection of the system board can reveal toasted parts,
green scum indicating corrosion on connections, dust
accumulation on cooling fins, bulging capacitors, etc. I've seen
drive cables that would fall off the connector under their own weight.
Periodically checking cables and memory for secure connections
doesn't take long.
Checking and replacing the CMOS backup battery is easy unless the
battery is soldered to the system board.
My experience with linux and macOS (and very different from Windows)
has been that filesystem corruption not associated with some obvious
issue like a system crash or power problem is more often caused by
hardware than by software. Whenever a disk acts up I check cables
and run smartctl or vendor tests -- some vendors have accepted smartctl
output from a linux or macOS system when issuing warranty return
authorizations.
It also pays to look for common failure modes for the hardware you
use. Some models have well-known failure modes. Examples from
my experience are capacitors, power supplies, fans, ethernet (card or
onboard), and cables.
George N. White III
_______________________________________________ users mailing list -- users@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to users-leave@xxxxxxxxxxxxxxxxxxxxxxx Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@xxxxxxxxxxxxxxxxxxxxxxx
