Comment on ChatGPT fried my drive!?

y0din@lemmy.world ⁨2⁩ ⁨days⁩ ago

• y0din@lemmy.world ⁨1⁩ ⁨day⁩ ago

  One more hopefully happy update:

  Based on everything you’ve shown so far in the information you have given, the most probable cause is that the drive was formatted with T10 DIF / Protection Information enabled (PROTECT=1), and you are now accessing it through a controller path that does not support DIF.

  This is a very common failure mode with enterprise SAS drives and sg_format.

  What this means in practice:

  • PROTECT=1 = the drive was formatted with DIF Type 1
  • Logical blocks are no longer plain 512/4096 bytes (e.g. 520/528 instead)
  • The HBA + driver must explicitly support T10 PI
  • If the controller does not support DIF, the drive may:
    • Be detected
    • But fail all I/O
    • Appear “dead” even though it is healthy

  This is not bricking. It is a configuration mismatch.

  How to fix it (most reliable path)

  You need to connect the drive to a DIF-capable SAS HBA (LSI/Broadcom, same type as originally used if possible).
  Best option is to do this on the original hardware, even via a USB live Linux environment.

  Once the drive is on a T10-capable controller, reformat it with protection disabled.

  Example (this will ERASE the drive and might take a LONG time to complete):

  sudo sg_format –format –size=512 –fmtpinfo=0 –pfu=0 /dev/sdX

  Key flags:

  • –fmtpinfo=0 → disables DIF / PROTECT
  • –size=512 (or 4096 if you prefer standard 4K)
  • –pfu=0 (disables PROTECTION flag, your GPT forgot to include this which actually disables the protection)
  • Use the correct /dev/sdX

  After this completes and the system is power-cycled, the drive should behave like a normal disk again on non-DIF controllers.

  Important notes

  • sg_format alone almost never permanently damages SAS drives
  • This exact scenario happens frequently when drives are moved between controllers
  • Until tested on a DIF-capable HBA, there is no evidence of permanent failure

  If you cannot access a T10-capable controller, the drive may remain unusable on that system, but still be perfectly recoverable elsewhere.

  A case of a user with a similar problem that got it fixed after a new format (found it on Google):

  https://www.truenas.com/community/threads/drives-formatted-with-type-1-protection-eventually-lead-to-data-loss.86007/

  source
• rook@lemmy.zip ⁨2⁩ ⁨days⁩ ago

  Thank you for helping! Like I said I’m a complete beginner with little knowledge of all this, means a lot 🤗

  just so you know I connected the drive to my dell pc, so its just the one broken drive not all 6.

  Exact drive model: SEAGATE ST4000NM0023 XMGG

  HBA model and firmware: lspci | grep -i raid 00:17.0 RAID bus controller: Intel Corporation SATA Controller [RAID mode] Its an LSI card Bought it here

  Kernel version / distro: I was using Truenas when I formatted it. Now trouble shooting on other PC got (6.8.0-38-generic), Linux Mint 22

  Whether the controller supports DIF/DIX (T10 PI): output of lspci -vv

  Whether other identical drives still work in the same slot/cable: yes all the other 5 drives worked when i set up a RAIDZ2 and a couple of them are exact same model of HDD

  COMMANDS This is what I got for each command: verbatim output from

  I really appreciate your knowledge and help 🙂
  Let me know if anything else is needed

  source

  • y0din@lemmy.world ⁨1⁩ ⁨day⁩ ago

    Thanks for the additional details, that helps, but there are still some critical gaps that prevent a proper diagnosis.

    Two important points first:

    The dmesg output needs to be complete, from boot until the moment the affected drive is first detected.
    What you posted is cut short and misses the most important part: the SCSI/SAS negotiation, protection information handling, block size reporting, and any sense errors when the kernel first sees the disk.

    Please reboot, then run as root or use sudo:

    dmesg -T > dmesg-full.txt

    1.  Do not filter or truncate it. Upload the full file.

    2. All diagnostic commands must be run with sudo/root, otherwise capabilities, mode pages, and protection features may not be visible or may be incomplete.

    Specifically, please re-run and provide full output (verbatim) of the following, all with sudo or as root, on the problem drive and (if possible) on a working identical drive for comparison:

    sudo lspci -nnkvv

    sudo lsblk -o NAME,MODEL,SIZE,PHY-SeC,LOG-SeC,ROTA

    sudo fdisk -l /dev/sdX

    sudo sg_inq -vv /dev/sdX

    sudo sg_readcap -ll /dev/sdX

    sudo sg_modes -a /dev/sdX

    sudo sg_vpd -a /dev/sdX

    Replace /dev/sdX with the correct device name as it appears at that moment.

    Why this matters:

    • The Intel SATA controller you listed is not the LSI HBA. We need to see exactly which controller the drive is currently attached to and what features the kernel believes it supports.

    • That Seagate model is a 520/528-capable SAS drive with DIF/T10 PI support. If it was formatted with protection enabled and is now attached to a controller/driver path that does not expect DIF, Linux will report I/O errors even though the drive itself is fine.

    • sg_format -vv output alone does not tell us the current logical block size, protection type, or mode page state.

    Important clarification:

    • Formatting the drive under TrueNAS (with a proper SAS HBA) and then attaching it to a different system/controller is a very common way to trigger exactly this situation.

    • This is still consistent with a recoverable configuration mismatch, not a permanently damaged disk.

    Once we have:

    • Full boot-time dmesg

    • Root-level SCSI inquiry, mode pages, and read capacity

    • Confirmation of which controller is actually in use

    …it becomes possible to say concretely whether the drive needs:

    • Reformatting to 512/4096 with protection disabled

    • A controller that supports DIF

    • Or if there is actual media or firmware failure (less likely)

    At this point, the drive is “unusable”, not proven “bricked”. The missing data is the deciding factor.

    One more important thing to verify, given the change of machines:

    Please confirm whether the controller in the original TrueNAS system is the same type of LSI/Broadcom SAS HBA as the one in the current troubleshooting system.

    This matters because:

    DIF/T10 PI is handled by the HBA and driver, not just the drive.

    A drive formatted with protection information on one controller may appear broken when moved to a different controller that does not support (or is not configured for) DIF.

    Many onboard SATA/RAID controllers and some HBAs will enumerate a DIF-formatted drive but fail all I/O.

    If the original TrueNAS machine used:

    • A proper SAS HBA with DIF support

    then the best recovery path may be to put the drive back into that original system and either:

    • Reformat it there with protection disabled, or

    • Access it normally if the controller and OS were already DIF-aware

    If the original controller was different:

    • Please provide lspci -nnkvv output from that system as well (using sudo or run as root)

    • And confirm the exact HBA model and firmware used in the TrueNAS SAS controller 

    At the moment, the controller change introduces an unknown that can fully explain the symptoms by itself. Verifying controller parity between systems is necessary before assuming the drive itself is at fault.

    source

    • rook@lemmy.zip ⁨20⁩ ⁨hours⁩ ago

      Thanks for the continued support! ❤

      I’ve attached an identical Segate SAS drive from the server.

      To confirm, it is the same LSI card that was in the TrueNAS server. I pulled it out of the server and put it into the trouble shooting machine, where I run the commands.

      It is this one: 01:00.0 Serial Attached SCSI controller [0107]: Broadcom / LSI SAS2308 PCI-Express Fusion-MPT SAS-2 [1000:0087] (rev 05)

      I did not see your other reply lol, I will also try this command that you recommended:

      sudo sg_format –format –size=512 –fmtpinfo=0 –pfu=0 /dev/sdb

      Also, the sg_format ran for less than 5 minutes, very quick. However, if I can recall, it did say it was completed.

      **Note: ** “Bricked Drive” turned to sdb

      Identical working drive installed as sda

      Here is the dmesg -T > dmesg-full.txt with the identical drive

      Here is the code from: (with the output for each drive, separately)

      sudo lspci -nnkvv

      sudo lsblk -o NAME,MODEL,SIZE,PHY-SeC,LOG-SeC,ROTA

      sudo fdisk -l /dev/sdX

      sudo sg_inq -vv /dev/sdX

      sudo sg_readcap -ll /dev/sdX

      sudo sg_modes -a /dev/sdX

      sudo sg_vpd -a /dev/sdX

      Thanks again for all the help, I await your reply. :)

      I will let you know the results of (sudo sg_format –format –size=512 –fmtpinfo=0 –pfu=0 /dev/sdb), as soon as it’s done.

      source

      • y0din@lemmy.world ⁨15⁩ ⁨hours⁩ ago

        Thanks for the update, that’s helpful.

        Confirming that the controller is a Broadcom / LSI SAS2308 and that it’s the same HBA that was used in the original TrueNAS system removes one major variable. It means the drive is now being tested under the same controller path it was previously attached to.

        The device mapping you described is clear:

        sda = known-good identical drive

        sdb = the problematic drive

        Running:

        sudo sg_format --format --size=512 --fmtpinfo=0 --pfu=0 /dev/sdb

        as you did is the correct next step to normalize the drive’s format and protection settings.

        A few general notes while this is in progress:

        • Some drives report completion before all internal states are fully settled, which will cause reduced performance as the operation continues until finished in the background
        • A power cycle after completion is recommended before testing the drive again

        At this point it makes sense to pause any further investigation until the current sg_format has fully completed and the system has been power-cycled.

        Once that’s done, the next step will be a direct comparison between sdb and the known-good sda using:

        sudo sg_readcap -lla

        • Reported logical and physical sector sizes

        • Protection / PI status

        As a general note going forward: on Linux / FreeBSD it’s safer to reference disks by persistent identifiers (e.g. /dev/disk/by-id/ on Linux or glabel on FreeBSD) rather than /dev/sdX, as device names can change across boots or hardware reordering.

        Post the results when you’re ready and the sg_format complete and we can continue from there.

        source

        • -> View More Comments
  • y0din@lemmy.world ⁨1⁩ ⁨day⁩ ago

    Sorry for the mess with replies.

    Please see the last comment I made on my own comment instead of yours where it should have been.

    Your drive is almost certainly not dead, you just need a T10 / DIF capable controller to disable the PROTECTION flag.

    Read more in the other post.

    source