becaus hd sntinel is rubbish (where you got the error about bad drive).
How long does it take for a dead HD to die?
- Thread starter Fritz
- Start date
Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
It is indeed the only result here. What we do know is that there are a lot of reallocated sectors and that is usually an indication that the drive is going bad, but without a constantly growing amount or remaps it's so hard to tell how long it would last, or if it re-allocated on a failed read or a failed write (a failed read would be the worst, that's just dataloss in most cases).
I can believe this but as far as GUI's go, what's a good alternative? I don't need 20 pages of details to scroll thru or a bunch of garbage that is irreverent and/or is unintelligible. Most programs can't see SAS drives or drives connect to a HBA or RAID controller.
smartctl can even poll SAS/SATA drives on SAS-expanders and nvme drives and you can choose what part of SMART-data you want to see - or just pipe its output to grep or awk to just output *exactly* what you want.
I don't get the fuss about needing GUIs for everything, if all one wants/needs is some data which is best represented in plaintext anyways.
that's why there is an '-h' switch and a manpage... Most manpages (e.g. the one for smartctl) have an "EXAMPLES" section towards the end, where usually the most common use cases are shown. 90% of the time you will find what you want to know within this short section, so no need to read the whole manpage (or just search in it via '/<keyword>')
and tbh - it can't take that long to memorize '-iA' to get the disk (i)nfo and SMART (A)ttributes.
Good point ...
Counterpoint: Shell alias (or just one-liner scripts)
Use case for shell alias: Every once in a while, I'm doing some disk/filesystem performance testing, and it is necessary/prudent to flush-and-clear the OS system buffers, to get a known and consistent base condition. But I can never remember the exact command(s) to do that.
Solution:
Code:
alias lobotomy='sync; echo 1 > /proc/sys/vm/drop_caches' # flush/clear buffers
alias fflobotomy='sync; echo 3 > /proc/sys/vm/drop_caches' # " " also metadata
======
"Life is such monotony,
Without a good lobotomy." --Roy Harper ; "Committed"
Last edited:
"How long does it take for a dead HD to die?"
This is obviously either a Zen koan or a 300-level quantum physics exam question. Or maybe it's both in a superimposed eigenstate.
This is obviously either a Zen koan or a 300-level quantum physics exam question. Or maybe it's both in a superimposed eigenstate.
Haven't fired up my dying HD server in a while. Just did. This wicked HD is now showing 100% health and more than 1000 days remaining life in HD Sentinel. When I first fired the server up it wouldn't show up in on the network so I pulled it out of the rack so I could hook a monitor up to it. It wasn't posting. I cut power, pulled out all the drives and tried again. She started normally with no hints in the Windows logs as to what happened. I then started inserting the drives one by one. When I got to this drive I was half way expecting the server to croak when I inserted it but it didn't. I had HD Sentinel open and waited for the drive to show up. When it did it showed 100% health. I have no idea what's going on here.
Yea, I'm thinking it's been the problem all along. I have another HD whose health improved for no reason. Going from 0% health to 100% is pretty bizarre.