SnapRAID SMART HDD test = Reliable?

    • OMV 5.x (beta)
    • SnapRAID SMART HDD test = Reliable?

      Hi all,

      I came across SnapRAID smart function with SnapRAID plugin.
      Please refer the picture attached.
      As the picture indicates, at (1), (2) & (3) = 2 hard disk will failed, probability = 84%
      I have started to a bit worry.......... ;(

      Questions:

      1) How reliable is SnapRAID smart information ?
      I used a freeware from CrystalDiskInfo, to test all my hard disks whether they are worthy
      to be used for long term data storage.

      I have tested all hard disk with CrystalDiskInfo, BEFORE using my OMV5 machine.
      All disks tested = GOOD, except 1 disk tested = CAUTION.

      So I had already removed the disk which indicated: CAUTION,
      and installed a good one.

      (Reference: harddrivegeek.com/crystaldiskinfo-caution/ )

      But to my surprise, SnapRAID smart give a different outlook of my hard disk's health.

      2)If the information provided is reliable, I need to be caution and stand by 2 hard disk, in case 2 hard disks do fail?

      Anyone got any comment/suggestion/advice?

      Thank you. :)

    • I have no idea how these failure rate probabilities are calculated. But I have seen numbers similar to what you are seeing over the last five years.

      So far no drives have failed or had any problems that would indicate a potential failure.

      I just keep adding disks and keep going. I have 12 disks: 9 hard disks, two SSDs, and one USB stick connected to my OMV machine. I have room for five more disks in the DAS box connected to the OMV box.
      --
      Google is your friend and Bob's your uncle!

      RAID - Its ability to disappoint is inversely proportional to the user's understanding of it.

      ASRock Rack C2550D4I C0 Stepping - 16GB ECC - Silverstone DS380 + Silverstone DS380 DAS Box.
    • gderf wrote:

      I have no idea how these failure rate probabilities are calculated. But I have seen numbers similar to what you are seeing over the last five years.

      So far no drives have failed or had any problems that would indicate a potential failure.

      I just keep adding disks and keep going. I have 12 disks: 9 hard disks, two SSDs, and one USB stick connected to my OMV machine. I have room for five more disks in the DAS box connected to the OMV box.
      Question is can I trust the data given by SnapRAID smart?

      Or the data given is over exaggerated comparing with the test made using CrystalDiskinfo.

      Did you try it on your system and see whether your 9 disks are currently in good health?

      Would you mind post a picture it on this thread showing your result?
    • ness1602 wrote:

      I always recommend adding hddsentinel to your box,and regularly executing it and monitoring disks.
      I use hddsentinel on Windows, can it work on Linux too? Is it a daemon and sends regular reports, or you have to access it via a web gui?
      EDIT: Found on the HDDsentinel website!!! Great, it can send an automated script everyday! Thanks for the hint buddy!

      @wepee have you enabled SMART reporting from OVM itself? That one is more detailed.
      I have a friend with a very used drive (2 bad unrecoverable bad sectors and 1300h uptime) which is rated by SNAPRAID at 99% failure, lol.
      My friend is aware of it and receives a daily SMART email from OMV, which monitors constantly the drive. If the health will degrade, he'll swap the drive.
      /// OMV BUILD IN PROGRESS - MY NAS KILLER /// omv 5.x + omvextrasorg

      i3 8300 - ASRock H370M-ITX/ac - 8GB RAM - Sandisk Ultra Flair 32GB (OMV), 256GB NVME SSD (Docker), 3x4TB HDD (Data) - Node 304 - Be quiet! Pure Power 11 350W

      The post was edited 1 time, last by thedarkness ().

    • wepee wrote:

      Question is can I trust the data given by SnapRAID smart?
      Or the data given is over exaggerated comparing with the test made using CrystalDiskinfo.

      Did you try it on your system and see whether your 9 disks are currently in good health?

      Would you mind post a picture it on this thread showing your result?
      I trust the data given by SnapRAID SMART. If and when it detects a real problem with a disk, I expect it to properly report it. I don't understand the failure probability reports. The numbers have been increasing since I started using OMV - more than five years ago, but I'm not seeing any SMART data that indicates any problems with any of my disks.

      My OMV machine runs headless and it can't run Windows programs like CrystalDiskinfo.
      --
      Google is your friend and Bob's your uncle!

      RAID - Its ability to disappoint is inversely proportional to the user's understanding of it.

      ASRock Rack C2550D4I C0 Stepping - 16GB ECC - Silverstone DS380 + Silverstone DS380 DAS Box.
    • thedarkness wrote:

      I have a friend with a very used drive (2 bad unrecoverable bad sectors and 1300h uptime) which is rated by SNAPRAID at 99% failure, lol.
      My friend is aware of it and receives a daily SMART email from OMV, which monitors constantly the drive. If the health will degrade, he'll swap the drive.
      @thedarkness Thank you for sharing your finding.

      In my opinion, I guess it is just a false alarm, or some sort of weird reporting / hard failure prediction that is probably not
      updated to reflect the newer hard disk drive's firmware/ that developer Andrea has forgotten to update.

      Anyway, like @'gderf said...... -go and read the manual, which I did.
      Click control + F, type in "smart" and I found this information from
      the manual.


      But when I dug in more information from the web, I found this: sourceforge.net/p/snapraid/dis…/1677233/thread/d53e8135/
      The guy has a similar problem.

      Below is the smart monitoring status within OMV.

      Both hard disks indicated with an arrow which SnapRAID smart reported to be 84% failure.
      Serial Nos. ending with last 4 digits: 5553 and 4648
      Both hard disk shows in good health.

    • wepee wrote:

      thedarkness wrote:

      I have a friend with a very used drive (2 bad unrecoverable bad sectors and 1300h uptime) which is rated by SNAPRAID at 99% failure, lol.
      My friend is aware of it and receives a daily SMART email from OMV, which monitors constantly the drive. If the health will degrade, he'll swap the drive.
      @thedarkness Thank you for sharing your finding.
      In my opinion, I guess it is just a false alarm, or some sort of weird reporting / hard failure prediction that is probably not
      updated to reflect the newer hard disk drive's firmware/ that developer Andrea has forgotten to update.
      Great :)
      Just keep drives under monitoring, and everything will be under control.
      /// OMV BUILD IN PROGRESS - MY NAS KILLER /// omv 5.x + omvextrasorg

      i3 8300 - ASRock H370M-ITX/ac - 8GB RAM - Sandisk Ultra Flair 32GB (OMV), 256GB NVME SSD (Docker), 3x4TB HDD (Data) - Node 304 - Be quiet! Pure Power 11 350W

      Post by gderf ().

      This post was deleted by the author themselves ().
    • With standard SMART testing (I do short tests), and with SMART notifications activated, the follow are the stat's that I pay attention to for spinning drives:
      (This info was gleaned from the backblaze study.)

      SMART 5 – Reallocated_Sector_Count.
      SMART 187 – Reported_Uncorrectable_Errors.
      SMART 188 – Command_Timeout.
      SMART 197 – Current_Pending_Sector_Count.
      SMART 198 – Offline_Uncorrectable.

      A "1" count in any stat, in itself, doesn't mean anything. But if these counts start incrementing upward, I'd consider swapping out the drive.
      _______________________________________________

      The rest of the available SMART stat's are OEM dependent, where they may be implemented differently from vendor to vendor. The only way to interpret them is to go on the OEM's website to see what they are and what triggers a count event. As an example, the raw counts on a Seagate (1 - Raw Read Error rate and 7 - seek error rate,) may be quite high but functionally, with built in error correction, these counts may mean nothing.

      As a side note:
      SMART 199 - UltraDMA CRC errors
      Are usually hardware connections or cable related.

      The post was edited 1 time, last by crashtest: edit ().

    • A 1 count in any of those five stats, for me, would trigger an order for a new disk.
      --
      Google is your friend and Bob's your uncle!

      RAID - Its ability to disappoint is inversely proportional to the user's understanding of it.

      ASRock Rack C2550D4I C0 Stepping - 16GB ECC - Silverstone DS380 + Silverstone DS380 DAS Box.