Replacing failing drive

[jrmacart]

I have 4 * 6 TB Drives in my array ( 3 data + 1 parity ). One of my data reported an error this morning...

 

I have a replacement 8 coming. I know the parity has to be the >= the largest data drive. But what will happen if it's not?

 

If I add the 8 tb as a data drive, will everything still work and only 6 tb will be usable or will it block me from adding it to the pool?

 

I know I should make the new drive my parity, but I don't really want to risk rebuilding my parity disk with an unhealthy drive.

[trurl]

Go to Tools-diagnostics and attach the complete Diagnostics ZIP file to your NEXT post in this thread. 

[jrmacart]

As requested sir.

 

Thanks.

edit: the problem disk is disk 1

edit2: remove diagnostics

 

Edited October 21, 2020 by jrmacart

[itimpi]

You will not be able to add the 8TB drive as a parity drive.  It is always possible that the disk in question has not really failed - I would wait for feedback on that as that would simplify things.

 

If the drive really has failed then in such a scenario one should use the Parity Swap procedure which is covered here in the online documentation.

[trurl]

Does look like an actual disk problem.

Oct 21 02:33:53 Tower kernel: ata1.00: exception Emask 0x0 SAct 0x4 SErr 0x0 action 0x0
Oct 21 02:33:53 Tower kernel: ata1.00: irq_stat 0x40000008
Oct 21 02:33:53 Tower kernel: ata1.00: failed command: READ FPDMA QUEUED
Oct 21 02:33:53 Tower kernel: ata1.00: cmd 60/08:10:40:00:00/00:00:00:00:00/40 tag 2 ncq dma 4096 in
Oct 21 02:33:53 Tower kernel:         res 41/40:00:40:00:00/00:00:00:00:00/00 Emask 0x409 (media error) <F>
Oct 21 02:33:53 Tower kernel: ata1.00: status: { DRDY ERR }
Oct 21 02:33:53 Tower kernel: ata1.00: error: { UNC }
Oct 21 02:33:53 Tower kernel: ata1.00: configured for UDMA/133
Oct 21 02:33:53 Tower kernel: sd 2:0:0:0: [sdf] tag#2 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=0x08
Oct 21 02:33:53 Tower kernel: sd 2:0:0:0: [sdf] tag#2 Sense Key : 0x3 [current] 
Oct 21 02:33:53 Tower kernel: sd 2:0:0:0: [sdf] tag#2 ASC=0x11 ASCQ=0x4 
Oct 21 02:33:53 Tower kernel: sd 2:0:0:0: [sdf] tag#2 CDB: opcode=0x88 88 00 00 00 00 00 00 00 00 40 00 00 00 08 00 00
Oct 21 02:33:53 Tower kernel: print_req_error: I/O error, dev sdf, sector 64
Oct 21 02:33:53 Tower kernel: md: disk1 read error, sector=0

 

You could run an extended SMART test but since you already have a larger replacement on the way I would just go ahead with the parity swap as linked by itimpi. That way you will be ready to upsize other disks in the future.

[jrmacart]

Thanks for the help. I'll proceed with the swap when the drive gets here next week.

 

 

[jrmacart]

Just an updat. Drive was swapped out and successfully rebuilt.

 

Once again, thanks for the help.