On 9/19/2025 10:20 PM, Tim wrote:
home user:
When I ordered the desktop, I asked that the drives be fully
S.M.A.R.T. monitorable.
& home user:
Years, ago, I was told by someone on this list that results of
S.M.A.R.T. monitoring of that hard drive were only partially
meaningful or applicable.
I suspect few shop staff know fine details of their products. They'd
look for "drive supports SMART" and that's probably it, and probably
few manufacturers provide much detail in their specs, either.
Predicting failure is a bit of a gamble. It can be sudden and by
surprise. There can be a cascading series of failures (bad blocks that
spread). There can be a rising temperature. And external factors can
be in force (rumbles and sudden bangs).
There's also the issue of thresholds. A manufacturer may believe their
drive can tolerate something more than someone else believes. Or they
can decide to raise the alarm very cautiously. A sceptic might say
that they do this just to sell you a new hard drive.
Also, there's the interpretation of RAW data. Some counts in the SMART
data may be literal (100 errors means that 100 errors have occurred),
but other things may be on a scale of 1 to 100, and requires
interpretation as to what's fine, cautious, bad, disaster levels.
Drive temperature can be like that, the drive *may* give a Celsius
reading, or a voltage reading across a thermistor that needs converting
to a temperature, and the SMART reader needs a look-up table for each
manufacturer, or even specific drives.
So yes, it's a tool in the bag, and it's up to you how much you trust
it. I'd probably keep a copy of the data from a brand new drive to
compare against over time, if you want to satisfy your curiosity.
To make use of SMART, you have to periodically check it. There's
smartd for that. And there has to be some way for you to see the results.
Logwatch was traditional (various things are checked daily and emailed to you).
There are other things that pop up notifications on the desktop (dunno what
happens if you're not looking at the time). And some people have some kind of
graphical system monitor always on the desktop showing some kind of system
health summaries (component temperatures, how full the drives are, etc).
There's perhaps one value you want to look at straight away, it's a
"conveyance" error number. It's supposed to mean some sort of value
related to how the drive was treated while being transported. All I
can think of it being some sort of shock sensor that doesn't reset
itself, so it doesn't need power to work. The idea being that if it's
bad, your drive has probably been dropped before you go it.
Good points and advice. Thank-you, Tim.
--
_______________________________________________
users mailing list -- [email protected]
To unsubscribe send an email to [email protected]
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/[email protected]
Do not reply to spam, report it:
https://pagure.io/fedora-infrastructure/new_issue
