Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.debian.user > #248354

Re: smartctl puzzlement new disk

Show all headers | View raw

[Multipart message — attachments visible in raw view] - view raw

On 14/05/2022 03:02, ghe2001 wrote:
> Supermicro workstation, Debian Buster, smartctl v 6.6
>
> I bought a new 12TB Western Digital Gold SATA disk the other day. 
> After testing it, smartctl says, among other things:
>
>  22 Unknown_Attribute       0x0023   001   001   025    Pre-fail 
> Always   FAILING_NOW 13
>
> And it looks like the test is aborted when smartctl sees that error.  
> And there's lots of yelling all over my monitor.
>
> I found some discussion of ID 22 on the 'Net -- it's often used to 
> mean 'helium level' which makes some sense on a 12TB disk.  But 
> testing the Seagate 12TB next to the Western Digital is OK -- sorta; 
> there's no ID 22 in the output.
>
> The Wikipedia article on S.M.A.R.T lists ID 22 as having to do with 
> the helium.
>
> On the web, it looked like there's a way of changing what ID 22's 
> called so smartctl won't hurl.  I'd like the test to just keep going 
> and tell me the other things it finds.
>
> My questions:
>
> Is this error on ID 22 anything to worry about?
> (I'm guessing that it isn't, and smartctl and/or something in the 
> Western Digital disk's firmware is mildly wrong)
>
> Is there really a way to change 22 so smartctl will report the helium 
> level?  How do I do that?

Regardless of whether smartctl knows what ID 22 means, the error is 
actually because the _value_ of ID 22 is too low.

SMART entries show four values:

  * VALUE - This is a "normalised" value, between 0 and 255. The
    RAW_VALUE (shown as the last column) is scaled according to the
    manufacturer's algorithm such that 100 is the ideal value.
  * WORST - This represents the worst level of VALUE that the drive has
    seen.
  * THRESH - This is a level which the manufacturer has determined. If
    VALUE is less than THRESH, there is a problem
  * RAW_VALUE - This is some internal value which the manufacturer knows
    the meaning of (sometimes it might be obvious such as the number of
    hours online, but in the case of the helium level it's not obvious
    what 13 represents)

So, Smartctl might not know that the helium level is low, but it _does_ 
know that ID 22 is at 1, which is less than 25, so therefore the drive 
is failing.

As far as the firmware is concerned, the drive is failing. Whether that 
decision is justified, only the manufacturer can really say.

>
> Is there a way to to get smartctl to finish the test?  If so, how do I 
> do that?
>
> TIA...
>
 >

Back to linux.debian.user | Previous | Next — Previous in thread | Find similar | Unroll thread

Thread

smartctl puzzlement new disk ghe2001 <ghe2001@pm.me> - 2022-05-14 04:30 +0200
  Re: smartctl puzzlement new disk David <bouncingcats@gmail.com> - 2022-05-14 04:40 +0200
    Re: smartctl puzzlement new disk ghe2001 <ghe2001@protonmail.com> - 2022-05-14 06:00 +0200
      Re: smartctl puzzlement new disk David <bouncingcats@gmail.com> - 2022-05-14 06:10 +0200
        Re: smartctl puzzlement new disk ghe2001 <ghe2001@protonmail.com> - 2022-05-14 21:20 +0200
          Re: smartctl puzzlement new disk David <bouncingcats@gmail.com> - 2022-05-15 04:10 +0200
  Re: smartctl puzzlement new disk David Christensen <dpchrist@holgerdanske.com> - 2022-05-14 05:10 +0200
  Re: smartctl puzzlement new disk Darac Marjal <mailinglist@darac.org.uk> - 2022-05-20 13:40 +0200

csiph-web