RAID1 degraded on NSA325 v2 after file deletion

13»

All Replies

  • Mijzelf
    Mijzelf Posts: 2,827  Guru Member
    250 Answers 2500 Comments Friend Collector Seventh Anniversary

    Yep. That I/O error definitely is the cause sdb is offline. And I guess there was a comparable error during sync, causing the sync to stop.

    The sync always stopped at 0.4%? That is about 1 minute? In that case it's probably not the power supply. If that is failing, it would stop almost directly when both disks are pulling their max power. And I think it's neither the SATA bus, as that doesn't know about 0.4%. So the disk itself remains.

    if you can suggest if I can / should upgrade something.

    I see you are using a rather old version of Tweaks. The latest contains a mitigation for a vulnerability which allows an attacker to take over the NAS when you only open a malicious webpage in the browser while being in the same LAN as the NAS. Scary. Yet I don't know if it's ever exploited.

  • siraph
    siraph Posts: 15  Freshman Member
    First Comment

    Hi! Happy new year!

    It's been a while, I didn't had much time to work on my NAS 'till today.

    Opened the NAS, extracted the "faulty" HD (disk 2), attached on an adapter and connected via USB to a Mac.

    Tried some diagnostic tools, starting from disk utility, and also trying e2fsck and badblocks, no errors found. badblocks kept running for a couple of hours, but had to stop it.

    Tried to initialize the disk with disk utils, MAC OS Journaled, and worked fine. Was able to write some files in the disk.

    I reinserted the disk into the NAS, initialized as a new volume using web interface, at the end it was shown as Healty, and could see it as a shared drive with Samba.

    Deleted the newly created volume, and now the disk was formatted in the right way to try to add it back to the RAID. Tried again:

    mdadm /dev/md0 --add /dev/sdb2

    But failed again.

    Attached to this message you can find the kmsg output.

    Thanks again for your help.

    kmsg20250119-3.txt

  • Mijzelf
    Mijzelf Posts: 2,827  Guru Member
    250 Answers 2500 Comments Friend Collector Seventh Anniversary

    Well, that's clear:

    <3>end_request: I/O error, dev sdb, sector 19023104

    There is an I/O error on sdb, around 10GB from the start. Which can happen, but this case is strange. The disk controller is expected to insert a spare sector for sector 19023104, and increment the Reallocated_Sector_Ct. The I/O error still happens (due to timing), but next time it's gone.

    However, in your start post your Reallocated_Sector_Ct is 0, while you already had tried to repair the array several times. In your start post this disk has a 'Raw_Read_Error_Rate' of 3562. I wonder if the disk fails to read some internal database, in which he can find which spare sector can be inserted. And I don't know how to test that.

Categories

Consumer Product Help Center

EnglishTiếng ViệtРусскийไทย中文（台灣）