Hi all,

My NAS was running out of space so I have decided to replace a 500GB drive with a new 2TB one, same size and model of my 2 parity drives.

These are the steps I have followed:

• Force parity check

• Shut down

• Remove the 500GB and swap It with the 2TB

• Power on

• Select the new drive to replace the missing old One

• Start Array

Everything went fine and Unraid started the data rebuilding process on the new drive but stupidly enough I decided to go in my docker section and stop all the containers. The (poor) reasoning was that I didn't want the container to alter the data while rebuilding.

When I got back to the main page, the data rebuilding was interrupted, new disk in error state with emulated content.

I don't want to brake anything so I am asking what's the best way to proceed at this point.

Diagnostic attached.

Thank you!

jimminas-diagnostics-20251012-0918.zip

Edited October 12, 2025Oct 12 by J1mm1
Title typo

Stopping Docker was not the reason, disk3 (is that the replacement?) has connection issues.

Reseat/clean contacts/change cable.

Oct 12 02:02:01 JimmiNas kernel: ata5.00: failed command: WRITE FPDMA QUEUED
Oct 12 02:02:01 JimmiNas kernel: ata5.00: cmd 61/68:f0:38:62:e0/03:00:01:00:00/40 tag 30 ncq dma 446464 out
Oct 12 02:02:01 JimmiNas kernel:         res 43/84:01:01:4f:c2/00:00:00:00:00/00 Emask 0x10 (ATA bus error)
Oct 12 02:02:01 JimmiNas kernel: ata5.00: status: { DRDY SENSE ERR }
Oct 12 02:02:01 JimmiNas kernel: ata5.00: error: { ICRC ABRT }
Oct 12 02:02:01 JimmiNas kernel: ata5: hard resetting link
Oct 12 02:02:02 JimmiNas kernel: ata5: SATA link up 3.0 Gbps (SStatus 123 SControl 320)
Oct 12 02:02:02 JimmiNas kernel: ata5.00: configured for UDMA/133
Oct 12 02:02:02 JimmiNas kernel: sd 5:0:0:0: [sdf] tag#30 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=3s
Oct 12 02:02:02 JimmiNas kernel: sd 5:0:0:0: [sdf] tag#30 Sense Key : 0xb [current] 
Oct 12 02:02:02 JimmiNas kernel: sd 5:0:0:0: [sdf] tag#30 ASC=0x47 ASCQ=0x0 
Oct 12 02:02:02 JimmiNas kernel: sd 5:0:0:0: [sdf] tag#30 CDB: opcode=0x2a 2a 00 01 e0 62 38 00 03 68 00
Oct 12 02:02:02 JimmiNas kernel: I/O error, dev sdf, sector 31482424 op 0x1:(WRITE) flags 0x0 phys_seg 109 prio class 0
Oct 12 02:02:02 JimmiNas kernel: md: disk3 write error, sector=31482360
Oct 12 02:02:02 JimmiNas kernel: md: disk3 write error, sector=31482368
Oct 12 02:02:02 JimmiNas kernel: md: disk3 write error, sector=31482376
Oct 12 02:02:02 JimmiNas kernel: md: disk3 write error, sector=31482384
Oct 12 02:02:02 JimmiNas kernel: md: disk3 write error, sector=31482392
Oct 12 02:02:02 JimmiNas kernel: md: disk3 write error, sector=31482400
Oct 12 02:02:02 JimmiNas kernel: md: disk3 write error, sector=31482408
Oct 12 02:02:02 JimmiNas kernel: md: disk3 write error, sector=31482416

Edited October 12, 2025Oct 12 by Kilrah

Ok, it's strange because I have used exactly the same cables and connectors used for the previous disk that worked just fine for over a year.

I have now tried to swap cables with another drive that is still working fine, while disk3 remains unavailable.

At power on the disk seems to be mounted correctly, how can I proceed to remove the error state or at least to understand what's happening?

Thanks!

Oct 12 13:11:24 JimmiNas emhttpd: mounting /mnt/disk3
Oct 12 13:11:24 JimmiNas emhttpd: shcmd (48): mkdir -m 0666 -p /mnt/disk3
Oct 12 13:11:24 JimmiNas emhttpd: /sbin/blkid /dev/md3p1 2>&1
Oct 12 13:11:24 JimmiNas emhttpd: /dev/md3p1: UUID="2b1867d9-164a-4baa-80a6-b3520bf1c68c" BLOCK_SIZE="512" TYPE="xfs"
Oct 12 13:11:24 JimmiNas emhttpd: shcmd (49): mount -t xfs -o noatime,nodiscard,nouuid /dev/md3p1 /mnt/disk3
Oct 12 13:11:24 JimmiNas kernel: XFS (md3p1): Mounting V5 Filesystem 2b1867d9-164a-4baa-80a6-b3520bf1c68c
Oct 12 13:11:25 JimmiNas kernel: XFS (md3p1): Ending clean mount
Oct 12 13:11:25 JimmiNas emhttpd: shcmd (50): xfs_growfs /mnt/disk3
Oct 12 13:11:25 JimmiNas root: meta-data=/dev/md3p1             isize=512    agcount=17, agsize=29303158 blks
Oct 12 13:11:25 JimmiNas root:          =                       sectsz=512   attr=2, projid32bit=1
Oct 12 13:11:25 JimmiNas root:          =                       crc=1        finobt=1, sparse=1, rmapbt=0
Oct 12 13:11:25 JimmiNas root:          =                       reflink=1    bigtime=1 inobtcount=1 nrext64=0
Oct 12 13:11:25 JimmiNas root:          =                       exchange=0   metadir=0
Oct 12 13:11:25 JimmiNas root: data     =                       bsize=4096   blocks=488378638, imaxpct=25
Oct 12 13:11:25 JimmiNas root:          =                       sunit=0      swidth=0 blks
Oct 12 13:11:25 JimmiNas root: naming   =version 2              bsize=4096   ascii-ci=0, ftype=1, parent=0
Oct 12 13:11:25 JimmiNas root: log      =internal log           bsize=4096   blocks=57232, version=2
Oct 12 13:11:25 JimmiNas root:          =                       sectsz=512   sunit=0 blks, lazy-count=1
Oct 12 13:11:25 JimmiNas root: realtime =none                   extsz=4096   blocks=0, rtextents=0
Oct 12 13:11:25 JimmiNas root:          =                       rgcount=0    rgsize=0 extents

Edited October 12, 2025Oct 12 by J1mm1

25 minutes ago, J1mm1 said:

while disk3 remains unavailable.

If you mean disabled, this is expected. Once a disk gets disabled, it needs to be rebuilt, if the emulated disk is mounting and contents look correct, you can rebuild on top, recommend replacing the original cables first, though.

31 minutes ago, JorgeB said:

If you mean disabled, this is expected. Once a disk gets disabled, it needs to be rebuilt, if the emulated disk is mounting and contents look correct, you can rebuild on top, recommend replacing the original cables first, though.

That’s nice to read!

How do I manually trigger the rebuild though?

1 hour ago, J1mm1 said:

How do I manually trigger the rebuild though?

• Stop array

• Unassign disk3

• Start array to make Unraid ‘forget’ current assignment

• Stop array

• Assign disk3

• Start array to begin rebuild.

Thank you, that worked as expected 🙂