Hey guys, i've been running unraid for quite sometime, and i couldn't be happier, its been great.

 

My system info is.

Unraid V6.0-rc3

24bay norco chassis, 21disks

 

I tried to replace one of my existing 4tb drives (disk19) with a 6tb drive (which I've done numerous times before), but this time, when i restarted the array to start up the array expansion, it started the rebuild, but when i checked on it a couple hours later, the rebuild stopped, and the newly replaced disk has a gray triangle on it, and my 'Parity status' is 'Data is invalid'. The array is started, and its serving data fine.

 

When i check the logs i see...

 

****

 

Dec 17 11:22:38 UNRAID kernel: sd 9:0:9:0: [sdw] tag#4 CDB: opcode=0x8a 8a 00 00 00 00 00 f0 ee 7a 08 00 00 04 00 00 00

Dec 17 11:22:38 UNRAID kernel: blk_update_request: I/O error, dev sdw, sector 4042160648

Dec 17 11:22:38 UNRAID kernel: sd 9:0:9:0: [sdw] tag#5 UNKNOWN(0x2003) Result: hostbyte=0x04 driverbyte=0x00

Dec 17 11:22:38 UNRAID kernel: sd 9:0:9:0: [sdw] tag#5 CDB: opcode=0x8a 8a 00 00 00 00 00 f0 ee 76 08 00 00 04 00 00 00

Dec 17 11:22:38 UNRAID kernel: blk_update_request: I/O error, dev sdw, sector 4042159624

Dec 17 11:22:38 UNRAID kernel: md: disk19 write error, sector=4042164680

Dec 17 11:22:38 UNRAID kernel: md: md_do_sync: got signal, exit...

Dec 17 11:22:38 UNRAID kernel: md: disk19 write error, sector=4042164688

Dec 17 11:22:38 UNRAID kernel: sd 9:0:9:0: [sdw] tag#6 UNKNOWN(0x2003) Result: hostbyte=0x04 driverbyte=0x00

Dec 17 11:22:38 UNRAID kernel: md: disk19 write error, sector=4042164696

Dec 17 11:22:38 UNRAID kernel: sd 9:0:9:0: [sdw] tag#6 CDB: opcode=0x8a 8a 00 00 00 00 00 f0 ee 72 08 00 00 04 00 00 00

Dec 17 11:22:38 UNRAID kernel: blk_update_request: I/O error, dev sdw, sector 4042158600

Dec 17 11:22:38 UNRAID kernel: md: disk19 write error, sector=4042164704

Dec 17 11:22:38 UNRAID kernel: md: disk19 write error, sector=4042164712

Dec 17 11:22:38 UNRAID kernel: md: disk19 write error, sector=4042164720

Dec 17 11:22:38 UNRAID kernel: md: disk19 write error, sector=4042164728

Dec 17 11:22:38 UNRAID kernel: md: disk19 write error, sector=4042164736

Dec 17 11:22:38 UNRAID kernel: sd 9:0:9:0: [sdw] tag#7 UNKNOWN(0x2003) Result: hostbyte=0x04 driverbyte=0x00

Dec 17 11:22:38 UNRAID kernel: md: disk19 write error, sector=4042164744

Dec 17 11:22:38 UNRAID kernel: sd 9:0:9:0: [sdw] tag#7 CDB: opcode=0x8a 8a 00 00 00 00 00 f0 ee aa 08 00 00 04 00 00 00

Dec 17 11:22:38 UNRAID kernel: blk_update_request: I/O error, dev sdw, sector 4042172936

Dec 17 11:22:38 UNRAID kernel: md: disk19 write error, sector=4042164752

 

****

and the "disk19 write error, sector=....." continues for numerous pages.

 

it looks like Disk19 went bad during the expansion, what is my suggested next step here?

Am i safe to stop the array, remove 'disk19' and replace it with another 6tb drive?

Will it rebuild correctly?

 

Any help would be appreciated!

Did you test that drive with preclear or something before trying to use it for the rebuild? Post SMART for the drive.

Sadly no, i didn't run a pre-clear on it (i think i learned a valuable lesson about skipping the preclear).

 

i can't seem to run a SMART test on it, when i 'spin up all drives' every drive except disk19 goes 'green ball', but disk19 stays as a gray triangle.

 

When i try to run a SMART test on it, it says that it needs to be spun up (but disk19 will not spin up).

 

Have you checked the connections, power and sata, to the drive?

I'll stop the array tomorrow, and reseat the drive (its plugs into a backplane on the norco), and the other drives on that backplane appear to be fine.

 

 

Hey guys, just figured i'd update my issue, incase anyone has it in the future.

 

I stopped the array, reseated the drive, and started the array again. Unraid detected the drive, but it detected it as "Unformatted". So it rebuilt parity for the array, with a missing/blank Disk19.

 

I stopped the array rebuild, replaced disk19 with a new disk, formatted disk19, and rebuilt the array again (which regenerated the parity with a blank disk19).

 

Everything appears to be fine, but i will have to copy the data from the old disk19 to the new disk19.

 

Thanks again for all the help!

rsbuc

Don't really understand this. I thought you were trying to rebuild a data disk, but you rebuilt parity?

That is correct, my initial plan was to replace an existing 4tb disk with a 6tb disk, but numerous write errors caused the expansion/rebuild to fail.

 

When i reseated the 6tb disk, i tried to restart the array with the same 6tb disk that had write errors, the array started with that drive showing "unformatted" and it started to rebuild the parity disk for the array with disk19 "missing" (since it was showing unformatted).

 

I didnt catch it until it already started to rebuild parity.

 

So i stopped the array, replaced disk 19, with a fresh 6tb disk, formatted it, and started the array (parity) rebuild.

Still unclear how you got to this point. With a replacement disk, the original disk is still being emulated, and it should build that emulated data to the replacement. If the emulated disk was unformatted, that would probably mean filesystem corruption. But it still wouldn't rebuild parity unless you did a new config. I wonder if your original disk will also seem unformatted. Nothing for it now but maybe you had other options had you asked for help. Hope you recover your data from the original.