I had a 1tb drive fail earlier this week and planned to replace it was a 14tb drive.  I've been running a preclear on it for the last few days and it just finished. 

 

I thought that the 1tb drive was still assigned to disk 1 so I assigned the 14tb drive to disk 5 and started the array.  A preclear started to run again but with a much shorter expected amount of time.  I realized that I needed to assign the 14tb drive to disk 1, so I stopped the preclear and array.  I'm no longer able to assign the 14tb drive to disk 1 though.

 

What is the best path forward to fully remove the 1tb drive, and add the 14tb drive to the array?

tower-diagnostics-20221208-0805.zip

Edited December 8, 20223 yr by faulty_lamp
added diagnostics

Unassign disk5, start array, stop array, now you can use the disk to replace disk1.

On 12/8/2022 at 10:38 AM, JorgeB said:

Unassign disk5, start array, stop array, now you can use the disk to replace disk1.

 

I was able to do this and after starting the array again a parity check began.  The parity check completed but the drive was blank under size.  I stopped the array again and then started it just now and it looks like it is running another parity check.

The icon next to the new 14tb disk is blue rather than green though.  I also see the same drive listed under historical devices.

 

Any idea what I'm doing wrong or what I need to do so that I can begin using this drive?

 

 

attached updated diagnostics

tower-diagnostics-20221209-1530.zip

11 hours ago, faulty_lamp said:

I was able to do this and after starting the array again a parity check began. 

You didn't do that, you did a new config (Tools -> New config) and in doing so lost the ability to rebuild disk1.

5 hours ago, JorgeB said:

You didn't do that, you did a new config (Tools -> New config) and in doing so lost the ability to rebuild disk1.

Ah, I was thinking that was the only way to unassign the disk. What would be the appropriate next steps to be able to use the new drive at this point?

Do another new config with the array set as you want then start the array to begin a parity sync.

On 12/11/2022 at 4:15 AM, JorgeB said:

Do another new config with the array set as you want then start the array to begin a parity sync.

 

That seemed to work, but shortly after the parity sync completed I started to receive errors on the new drive and then it disabled itself.  I tried removing the drive and re-adding it, but it remains disabled.

 

I forgot to mention, but this is the first SAS drive that I'm adding to my array.  All of my SATA drives are hooked up directly to the mobo.  I am using an ASR-7805 SAS card in HBA mode with an SFF-8643 to SFF-8482 cable.

 

Whenever I first added the drive I ran a preclear and everything came back fine with it.  I've attached both the diagnostics and SMART report for the drive.

tower-diagnostics-20221214-1702.zip tower-smart-20221214-1702.zip

SMART looks OK and disk failed immediately after a write attempt:

 

Dec 13 07:33:06 Tower  emhttpd: shcmd (358): mkfs.xfs -m crc=1,finobt=1 -f /dev/md1
Dec 13 07:33:07 Tower kernel: sd 1:1:3:0: [sdb] tag#243 UNKNOWN(0x2003) Result: hostbyte=0x05 driverbyte=DRIVER_OK cmd_age=0s
Dec 13 07:33:07 Tower kernel: sd 1:1:3:0: [sdb] tag#243 CDB: opcode=0x2a 2a 00 60 00 08 10 00 00 40 00
Dec 13 07:33:07 Tower kernel: I/O error, dev sdb, sector 12884918400 op 0x1:(WRITE) flags 0x4000 phys_seg 64 prio class 0
Dec 13 07:33:07 Tower kernel: md: disk1 write error, sector=12884918336
Dec 13 07:33:07 Tower kernel: md: disk1 write error, sector=12884918344

 

See if you can format the disk using UD, SAS disks sometimes have some kind of write protection enabled.

On 12/15/2022 at 3:52 AM, JorgeB said:

SMART looks OK and disk failed immediately after a write attempt:

 

Dec 13 07:33:06 Tower  emhttpd: shcmd (358): mkfs.xfs -m crc=1,finobt=1 -f /dev/md1
Dec 13 07:33:07 Tower kernel: sd 1:1:3:0: [sdb] tag#243 UNKNOWN(0x2003) Result: hostbyte=0x05 driverbyte=DRIVER_OK cmd_age=0s
Dec 13 07:33:07 Tower kernel: sd 1:1:3:0: [sdb] tag#243 CDB: opcode=0x2a 2a 00 60 00 08 10 00 00 40 00
Dec 13 07:33:07 Tower kernel: I/O error, dev sdb, sector 12884918400 op 0x1:(WRITE) flags 0x4000 phys_seg 64 prio class 0
Dec 13 07:33:07 Tower kernel: md: disk1 write error, sector=12884918336
Dec 13 07:33:07 Tower kernel: md: disk1 write error, sector=12884918344

 

See if you can format the disk using UD, SAS disks sometimes have some kind of write protection enabled.

 

That looks like that worked, thank you for all of the assistance!