Odd IDE failure and SMART messages

From: Cameron Heide (cnh_pub_at_shaw.ca)
Date: 12/30/03 

Date: Tue, 30 Dec 2003 19:56:16 GMT

One of my systems was left hung a while ago and shortly before the
failure I found the following messages from smartd in the syslog:

Dec 20 22:17:50 derzon -- MARK --
Dec 20 22:17:55 derzon smartd[716]: Device: /dev/hda, SMART Prefailure
Attribute: 7 Seek_Error_Rate changed from 200 to 100
Dec 20 22:17:56 derzon smartd[716]: Device: /dev/hdb, SMART Prefailure
Attribute: 7 Seek_Error_Rate changed from 200 to 100
Dec 20 22:17:56 derzon smartd[716]: Device: /dev/hdd, SMART Prefailure
Attribute: 7 Seek_Error_Rate changed from 200 to 100
Dec 20 22:37:50 derzon -- MARK --
Dec 20 22:57:50 derzon -- MARK --
Dec 27 18:03:42 derzon syslogd 1.4.1: restart.

It strikes me as being a bit odd that all three drives would experience
the exact same kind of failure at the same time. The only thing that
immediately comes to mind is some kind of fluctuation from the power
supply, perhaps.

Any other ideas as to what might have happened or what this means?
Thanks!

(Misc. relevant info: Kernel 2.6.0, Slack 9.1 distro, all three
drives are WD 1200BB Caviar drives in a software RAID-5 set,
smartmontools version 5.21, Gigabyte 7DXR motherboard...)

Relevant Pages

  • Re: Odd IDE failure and SMART messages
    ... > Unfortunately after restarting the system, the drives aren't telling me ... the worst value and the third is the failure ... contrary to what the SYSLOG messages say. ... I'm concerned that this might be a bug in smartd. ...
    (comp.os.linux.hardware)
  • Re: [OT] Smartd: Worth the effort?
    ... SMARTd doesn't identify anything. ... In my experience, mostly with SCSI drives, they always were smart enough ... to whine loudly before failure: ...
    (Fedora)
  • Re: Imminent drive failure?
    ... >> It is normal for disk blocks to go bad, ... >> disk drives handle them automatically, ... inclined to treat this drive/setup as indicative of "failure". ... Even if it passes all tests, I would definitely run smartd. ...
    (linux.redhat)
  • Re: Solid capacitors vs electrolytic...anyone with an opinion either way?
    ... found no correlation between power cycle counts and failures. ... Since IIRC modern disks are designed to survive 50K power cycles, one would not expect to see any noticeable failure rate when studying drives that only underwent a few tens or at most hundreds of power cycles during their lifetime. ...
    (comp.arch)
  • Re: [opensuse] Re: Hard Disk Failing
    ... increase in failure rate if the drive has light usage, ... pattern of the quaintly labelled 'infant mortality' in which drives are ... report on survival rates after the first event but do not report on ...
    (SuSE)