[smartmontools-support] Reports Health Status OK? But No SMART Support Available?

[Schorschi Decker]

Ah, cool.  Thanks for the clarification.

I have worked with SCSI based technology for most of my 30 year IT 
career.  But, I had not used SmartMonTools until recently. Typical in 
enterprise monitoring solutions, the hardware status (reporting) is 
layered, filtered, or otherwise presented via vendor specific vendor 
agents, SNMP, etc., so you don't usually get to the (raw) registers, 
attributes or low-level state information.  After all, when you are 
monitoring 100K servers world wide, and 1 disk in a RAID 5 configuration 
goes from Green to Red, and the Hot Spare spins up,  for example, you 
never worry about the specific cause (even if it is a false positive), 
you just schedule the suspect drive for replacement, by local on-site 
support, right?

I remember one situation years ago, where a recent disk firmware update 
caused 10s of drives on 1000s servers, marked good drives, as in a 
potential failure state, it was a software bug, that got through vendor 
code QA/QC.  The chaos and confusion it caused, was frankly, nuts, and 
vendor definitely had egg on face for that one!

-DD

On 3/16/2022 23:45, Claudio Kuenzler wrote:
>
>
> On Thu, Mar 17, 2022 at 7:36 AM Schorschi Decker <schorschi at dc.rr.com> 
> wrote:
>
>     I asked the question... because nothing typically reported was
>     reported.  Not even status determined by attribute scan.  This is
>     the first time I have ever seen SmartMonTools not report results
>     in sufficient detail to be clear on how the result was achieved,
>     hence the question.
>
>
> Fair question. The smartctl output might look weird when you're used 
> to ATA devices and their SMART attributes. But if you've worked with 
> SCSI disks a while ago, this is kind of normal behaviour.
> How exactly the Health of the drive is assessed, I can't tell you. But 
> the health status (mostly) works. See 
> https://www.claudiokuenzler.com/blog/411/check_smart-saves-server-lifes-defect-hard-drive-detected 
> for such a case. You can also see the "Elements in grown defect list" 
> on this article as these were all SCSI drives.
>
>     Moreover, the comment 'Never assume 100% accuracy on the reported
>     data though' is what?  Is that a blanket disclaimer or a comment
>     specific to IDE device status determination?
>
> That is a general disclaimer :-). The data retrieved and shown by 
> SMART should be used as helpful indicator. But never fully rely on it. 
> A disk not indicating any errors in any attribute can still fail out 
> of the blue.
>
> _______________________________________________
> Smartmontools-support mailing list
> Smartmontools-support at listi.jpberlin.de
> https://listi.jpberlin.de/mailman/listinfo/smartmontools-support
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://listi.jpberlin.de/pipermail/smartmontools-support/attachments/20220317/6c120a41/attachment.htm>