Raid 5 after extension: clean, degraded

1. offizieller Beitrag

    Hi all,


    I'm on 6.0.28-3 (Shaitan). I extended my RAID-5 to a new 4 TB HDD, having 3x 4 TB already in the system. During extension, the system shut down (for some unknown reason, maybe overheating), but when I started it again, it continued with the extension. When it ended (seemingly successfully), I was in a hurry and just quickly extended the file system, which worked.


    Now having a closer look at the RAID, it tells me it's in the state "clean, degraded":

    Code
    Version : 1.2 Creation Time : Sun Sep 2 04:05:15 2018 Raid Level : raid5 Array Size : 11720661504 (11177.69 GiB 12001.96 GB) Used Dev Size : 3906887168 (3725.90 GiB 4000.65 GB) Raid Devices : 4 Total Devices : 3 Persistence : Superblock is persistent
Intent Bitmap : Internal
Update Time : Fri Jan 27 07:00:56 2023 State : clean, degraded Active Devices : 3 Working Devices : 3 Failed Devices : 0 Spare Devices : 0
Layout : left-symmetric Chunk Size : 512K

Consistency Policy : bitmap
Name : openmediavault:RAIDAR (local to host openmediavault) UUID : e98b7abd:4f328c81:40a102c3:1824afcf Events : 79896
Number Major Minor RaidDevice State 0 8 16 0 active sync /dev/sdb 1 8 32 1 active sync /dev/sdc 2 8 48 2 active sync /dev/sdd - 0 0 3 removed

    Googling suggested to me the mdadm --add command to add the missing drive back to the array. However, I would have expected the "recover" option in the GUI to do the same, but I cannot select a device there:



    Does anyone have experience with this? Can I safely execute the mdadm --add command or do I need to do something else?


    Here some detailed information:

    cat /proc/mdstat

    Code
    Personalities : [raid6] [raid5] [raid4] [linear] [multipath] [raid0] [raid1] [raid10]
md127 : active raid5 sdc[1] sdb[0] sdd[2]
11720661504 blocks super 1.2 level 5, 512k chunk, algorithm 2 [4/3] [UUU_]
bitmap: 20/30 pages [80KB], 65536KB chunk

    blkid

    Code
    /dev/sda1: UUID="64ae1488-3bd9-4236-8742-9ea44db6f56c" BLOCK_SIZE="4096" TYPE="ext4" PARTUUID="76aa5ac0-01"
/dev/sda5: UUID="c2b0cb47-aeec-4b5a-8285-857b1c56da54" TYPE="swap" PARTUUID="76aa5ac0-05"
/dev/sdb: UUID="e98b7abd-4f32-8c81-40a1-02c31824afcf" UUID_SUB="a36eadb0-2348-fb83-ec76-65c9fa5df48b" LABEL="openmediavault:RAIDAR" TYPE="linux_raid_member"
/dev/sdc: UUID="e98b7abd-4f32-8c81-40a1-02c31824afcf" UUID_SUB="94cf7512-43e5-3957-7060-0e6cc0cdd526" LABEL="openmediavault:RAIDAR" TYPE="linux_raid_member"
/dev/sdd: UUID="e98b7abd-4f32-8c81-40a1-02c31824afcf" UUID_SUB="f1ae8b96-55da-2541-bc00-7be870687109" LABEL="openmediavault:RAIDAR" TYPE="linux_raid_member"
/dev/md127: LABEL="Raidar" UUID="5d21dac9-d7ba-4831-9d29-e6d9d8de5b3b" BLOCK_SIZE="4096" TYPE="ext4"
/dev/sde: UUID="e98b7abd-4f32-8c81-40a1-02c31824afcf" UUID_SUB="e80184c3-5dc3-17b4-1f73-a6f95f5fb718" LABEL="openmediavault:RAIDAR" TYPE="linux_raid_member"
/dev/sdf1: UUID="b533ba9f-52ff-9d49-8092-a954a53881e4" BLOCK_SIZE="4096" TYPE="ext4" PTUUID="d433308c" PTTYPE="dos" PARTUUID="d433308c-01"

    fdisk -l | grep "Disk "

    cat /etc/mdadm/mdadm.conf

    mdadm --detail --scan --verbose

    Code
    ARRAY /dev/md/openmediavault:RAIDAR level=raid5 num-devices=4 metadata=1.2 name=openmediavault:RAIDAR UUID=e98b7abd:4f328c81:40a102c3:1824afcf
devices=/dev/sdb,/dev/sdc,/dev/sdd


    Any help is appreciated, thank you.

    • Offizieller Beitrag
    Zitat von Skullchuck

    However, I would have expected the "recover" option in the GUI to do the same, but I cannot select a device there:

    That is because the drive /dev/sde has a raid signature on it according to blkid


    I assume that /dev/sde is the drive you added to grow the array, if that's the case then mdadm --add /dev/md127 /dev/sde should add the drive back to the array

    Zitat von geaves

    That is because the drive /dev/sde has a raid signature on it according to blkid


    I assume that /dev/sde is the drive you added to grow the array, if that's the case then mdadm --add /dev/md127 /dev/sde should add the drive back to the array

    Thank you for the quick response. OK, I did that. Unfortunately it's still "clean, degraded":


    cat /proc/mdstat

    Code
    Personalities : [raid6] [raid5] [raid4] [linear] [multipath] [raid0] [raid1] [raid10]
md127 : active raid5 sde[3](F) sdc[1] sdb[0] sdd[2]
11720661504 blocks super 1.2 level 5, 512k chunk, algorithm 2 [4/3] [UUU_]
bitmap: 11/30 pages [44KB], 65536KB chunk

    Looks like the drive is faulty :(

    Do you know if this must be a hardware error or do I have any (software-wise) recovery options from here on?

    • Offizieller Beitrag
    Zitat von Skullchuck

    Looks like the drive is faulty

    Is the drive new or repurposed? Have you run a long SMART test on it? Might not be the drive could be, m'board connection, sata cable, intermittent power

    Zitat von Skullchuck

    Do you know if this must be a hardware error or do I have any (software-wise)

    This is hardware related

    Zitat von Skullchuck

    recovery options from here on

    A backup :) the current array is still accessible, but I always advise restricting access as much as possible

    The drive is brand-new. I can only image that the shutdown during extension broke something :(

    I would like to run a long SMART test, but again no device is listed:



    Am I doing something wrong?


    EDIT: Solved it by

    sudo smartctl -t long /dev/sde


    It will be running for 10 hours...

    • Offizieller Beitrag
    Zitat von Skullchuck

    The drive is brand-new

    Then you would not expect mdadm to mark it as failed, but then again nothing is guaranteed. If the SMART output shows nothing relevant then I would suggest a wipe of the drive Storage -> Disks -> Wipe and run secure


    If the drive fails again to be added to the array then you're looking at hardware, or if you're able to connect it to a Windows machine and run the manufacturers diagnostic tools


    To get an RMA on the drive you need to ensure that it is the drive and not something else

    So the test is still running, but curiously I looked at sudo smartctl -a /dev/sde and found some errors in the log:



    It doesn't seem like these error codes (800-804) are the official ones from WD, but maybe someone knows how to interpret them...

  • Skullchuck

    Hat das Label gelöst hinzugefügt.

Jetzt mitmachen!

Sie haben noch kein Benutzerkonto auf unserer Seite? Registrieren Sie sich kostenlos und nehmen Sie an unserer Community teil!

Benutzerkonto erstellen Anmelden