So i had a 2TB drive go out, I ran the array for 2-3 days while the new 3TB was being shipped. No issue at all with the server while unprotected (other than the fact it was unprotected).

 

The 3TB arrives, and I pop it in, it sees it and I begin the rebuild, all as expected, then the server crashes. Just straight vertical lines on the attached monitor, completely froze, drops off the network, KB unresponsive.

 

I power down and restart the server, up it comes and begins the rebuild, and does it again, this repeats. The server crashes between 1-3 hours into the rebuild. I read a similar post where LIMETECH says to open the log and then paste when it crashes, but it doesn't seem to be populating real-time. Below is what is in the log file right now. Any help would be appreciated.

 

/usr/bin/tail -n 40 -f /var/log/syslog

Sep 4 09:43:11 MotherShip kernel: md: disk1 read error, sector=286165472

Sep 4 09:43:11 MotherShip kernel: md: disk1 read error, sector=286165480

Sep 4 09:43:11 MotherShip kernel: md: disk1 read error, sector=286165488

Sep 4 09:43:11 MotherShip kernel: md: disk1 read error, sector=286165496

Sep 4 09:43:11 MotherShip kernel: md: disk1 read error, sector=286165504

Sep 4 09:43:11 MotherShip kernel: md: disk1 read error, sector=286165512

Sep 4 09:43:11 MotherShip kernel: md: disk1 read error, sector=286165520

Sep 4 09:43:11 MotherShip kernel: md: disk1 read error, sector=286165528

Sep 4 09:43:11 MotherShip kernel: md: disk1 read error, sector=286165536

Sep 4 09:43:11 MotherShip kernel: md: disk1 read error, sector=286165544

Sep 4 09:43:11 MotherShip kernel: md: disk1 read error, sector=286165552

Sep 4 09:43:11 MotherShip kernel: md: disk1 read error, sector=286165560

Sep 4 09:43:17 MotherShip kernel: ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0

Sep 4 09:43:17 MotherShip kernel: ata1.00: irq_stat 0x40000001

Sep 4 09:43:17 MotherShip kernel: ata1.00: failed command: READ DMA EXT

Sep 4 09:43:17 MotherShip kernel: ata1.00: cmd 25/00:f8:c8:42:0f/00:00:11:00:00/e0 tag 0 dma 126976 in

Sep 4 09:43:17 MotherShip kernel: res 51/40:17:a0:43:0f/00:00:11:00:00/e0 Emask 0x9 (media error)

Sep 4 09:43:17 MotherShip kernel: ata1.00: status: { DRDY ERR }

Sep 4 09:43:17 MotherShip kernel: ata1.00: error: { UNC }

Sep 4 09:43:17 MotherShip kernel: ata1.00: SB600 AHCI: limiting to 255 sectors per cmd

Sep 4 09:43:17 MotherShip kernel: ata1.00: SB600 AHCI: limiting to 255 sectors per cmd

Sep 4 09:43:17 MotherShip kernel: ata1.00: configured for UDMA/133

Sep 4 09:43:17 MotherShip kernel: sd 0:0:0:0: [sdb] Unhandled sense code

Sep 4 09:43:17 MotherShip kernel: sd 0:0:0:0: [sdb]

Sep 4 09:43:17 MotherShip kernel: Result: hostbyte=0x00 driverbyte=0x08

Sep 4 09:43:17 MotherShip kernel: sd 0:0:0:0: [sdb]

Sep 4 09:43:17 MotherShip kernel: Sense Key : 0x3 [current] [descriptor]

Sep 4 09:43:17 MotherShip kernel: Descriptor sense data with sense descriptors (in hex):

Sep 4 09:43:17 MotherShip kernel: 72 03 11 04 00 00 00 0c 00 0a 80 00 00 00 00 00

Sep 4 09:43:17 MotherShip kernel: 11 0f 43 a0

Sep 4 09:43:17 MotherShip kernel: sd 0:0:0:0: [sdb]

Sep 4 09:43:17 MotherShip kernel: ASC=0x11 ASCQ=0x4

Sep 4 09:43:17 MotherShip kernel: sd 0:0:0:0: [sdb] CDB:

Sep 4 09:43:17 MotherShip kernel: cdb[0]=0x28: 28 00 11 0f 42 c8 00 00 f8 00

Sep 4 09:43:17 MotherShip kernel: end_request: I/O error, dev sdb, sector 286213024

Sep 4 09:43:17 MotherShip kernel: ata1: EH complete

Sep 4 09:43:17 MotherShip kernel: md: disk1 read error, sector=286212960

Sep 4 09:43:17 MotherShip kernel: md: disk1 read error, sector=286212968

Sep 4 09:43:17 MotherShip kernel: md: disk1 read error, sector=286212976

Sep 4 09:43:17 MotherShip kernel: md: disk1 read error, sector=286212984

What version unRAID? 4.7 does not support 3TB.

 

Did you pre-clear the new drive? While not strictly necessary when doing a drive rebuild, it can help test a new drive and weed out potential problems.

 

Also, is your parity 3Tb as well?

ver: 5.0-rc16c

 

It has a 3TB in there as parity and another 3TB drive in there for storage. I did not preclear as Limetech said wasn't required.

 

check power/data cables to the new drive.

Ok I put a new data cable and plugged in a different power cable off a different cable cluster.

 

Restarted now its at 8.1%... will update if this was the fix for future Limers with this issue.

 

 

Looks like you were right, one of the 2 was the culprit. I wonder of the previous drive really needed replacing. No biggie I gained a TB

 

Thanks for all your help!

 

R1

I have seen this numerous times. Plug the old supposedly failed drive back in and it will preclear without issue. I then preclear a few more times and then label it as a spare.

 

Cable connections and backplanes are much more liable to cause a fail than a drive.

 

Sent from my Nexus 4 using Tapatalk 4

 

will do!

i agree, triple pre-clear it and save as a spare if it passes.

can pre-clear function while array is up and running? I just ordered a 4 port sata controller to replace one of the 2 ports in there to make that drive available if pre-clear OKs it.

can pre-clear function while array is up and running? I just ordered a 4 port sata controller to replace one of the 2 ports in there to make that drive available if pre-clear OKs it.

There should be no problem pre-clearing drives that are not part of the array.  I do it regularly.