I had a disk fail last night.  I'm running 4.5.6.

 

Here is the part of the syslog when everything went bad:

 

Sep  3 20:35:59 Tower kernel: ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 frozen

Sep  3 20:35:59 Tower kernel: ata1.00: failed command: WRITE DMA EXT

Sep  3 20:35:59 Tower kernel: ata1.00: cmd 35/00:08:3f:00:50/00:00:20:00:00/e0 tag 0 dma 4096 out

Sep  3 20:35:59 Tower kernel:          res 40/00:ff:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)

Sep  3 20:35:59 Tower kernel: ata1.00: status: { DRDY }

Sep  3 20:35:59 Tower kernel: ata1: hard resetting link

Sep  3 20:36:04 Tower kernel: ata1: link is slow to respond, please be patient (ready=0)

Sep  3 20:36:09 Tower kernel: ata1: SRST failed (errno=-16)

Sep  3 20:36:09 Tower kernel: ata1: hard resetting link

Sep  3 20:36:14 Tower kernel: ata1: link is slow to respond, please be patient (ready=0)

Sep  3 20:36:19 Tower kernel: ata1: SRST failed (errno=-16)

Sep  3 20:36:19 Tower kernel: ata1: hard resetting link

Sep  3 20:36:25 Tower kernel: ata1: link is slow to respond, please be patient (ready=0)

Sep  3 20:36:54 Tower kernel: ata1: SRST failed (errno=-16)

Sep  3 20:36:54 Tower kernel: ata1: limiting SATA link speed to 1.5 Gbps

Sep  3 20:36:54 Tower kernel: ata1: hard resetting link

Sep  3 20:36:59 Tower kernel: ata1: SRST failed (errno=-16)

Sep  3 20:36:59 Tower kernel: ata1: reset failed, giving up

Sep  3 20:36:59 Tower kernel: ata1.00: disabled

Sep  3 20:36:59 Tower kernel: ata1.00: device reported invalid CHS sector 0

Sep  3 20:36:59 Tower kernel: ata1: EH complete

Sep  3 20:36:59 Tower kernel: sd 1:0:0:0: [sda] Unhandled error code

Sep  3 20:36:59 Tower kernel: sd 1:0:0:0: [sda] Result: hostbyte=0x04 driverbyte=0x00

Sep  3 20:36:59 Tower kernel: sd 1:0:0:0: [sda] CDB: cdb[0]=0x2a: 2a 00 20 50 00 3f 00 00 08 00

Sep  3 20:36:59 Tower kernel: end_request: I/O error, dev sda, sector 542113855

Sep  3 20:36:59 Tower kernel: sd 1:0:0:0: [sda] Unhandled error code

Sep  3 20:36:59 Tower kernel: sd 1:0:0:0: [sda] Result: hostbyte=0x04 driverbyte=0x00

Sep  3 20:36:59 Tower kernel: sd 1:0:0:0: [sda] CDB: cdb[0]=0x2a: 2a 00 20 50 9e c7 00 00 08 00

Sep  3 20:36:59 Tower kernel: end_request: I/O error, dev sda, sector 542154439

Sep  3 20:36:59 Tower kernel: sd 1:0:0:0: [sda] Unhandled error code

Sep  3 20:36:59 Tower kernel: sd 1:0:0:0: [sda] Result: hostbyte=0x04 driverbyte=0x00

Sep  3 20:36:59 Tower kernel: sd 1:0:0:0: [sda] CDB: cdb[0]=0x2a: 2a 00 00 00 00 bf 00 00 08 00

Sep  3 20:36:59 Tower kernel: end_request: I/O error, dev sda, sector 191

Sep  3 20:36:59 Tower kernel: sd 1:0:0:0: [sda] Unhandled error code

Sep  3 20:36:59 Tower kernel: sd 1:0:0:0: [sda] Result: hostbyte=0x04 driverbyte=0x00

Sep  3 20:36:59 Tower kernel: sd 1:0:0:0: [sda] CDB: cdb[0]=0x2a: 2a 00 20 58 00 3f 00 00 08 00

Sep  3 20:36:59 Tower kernel: end_request: I/O error, dev sda, sector 542638143

Sep  3 20:36:59 Tower kernel: sd 1:0:0:0: [sda] Unhandled error code

Sep  3 20:36:59 Tower kernel: sd 1:0:0:0: [sda] Result: hostbyte=0x04 driverbyte=0x00

Sep  3 20:36:59 Tower kernel: sd 1:0:0:0: [sda] CDB: cdb[0]=0x2a: 2a 00 20 5a e9 67 00 00 08 00

Sep  3 20:36:59 Tower kernel: end_request: I/O error, dev sda, sector 542828903

Sep  3 20:36:59 Tower kernel: sd 1:0:0:0: [sda] Unhandled error code

Sep  3 20:36:59 Tower kernel: sd 1:0:0:0: [sda] Result: hostbyte=0x04 driverbyte=0x00

Sep  3 20:36:59 Tower kernel: sd 1:0:0:0: [sda] CDB: cdb[0]=0x2a: 2a 00 59 1d 14 df 00 00 08 00

Sep  3 20:36:59 Tower kernel: end_request: I/O error, dev sda, sector 1495078111

Sep  3 20:36:59 Tower kernel: sd 1:0:0:0: [sda] Unhandled error code

 

Is this a bad cable or is this drive really toast?

 

Thanks in advance

We cannot tell from the log other than communications to the disk are timing out.

 

You can first try:

smartctl -d ata -a /dev/sda

and see if it responds at all.

 

If not...You can try:

stopping the array

Powering down

re-seating the power AND SATA cables

powering back up again.

 

This will NOT put the disk back on-line, but it might then respond to a

smartctl -d ata -a /dev/sda

 

If it then does respond:

stop the array

un-assign the drive

start the array with it un-assigned (to get it to forget the old model/serial number of the drive)

Stop the array once more

re-assign the drive  (it will then accept it as its own replacement)

Start the array once more.  It will then begin the process of re-constructing the failed drive from parity and the other drives.

 

Once re-constructed, the array will be back to normal.

 

If the drive is really dead:

Stop the array

Power down

Install the replacement disk

power up

Start the array.  (It will begin the process of re-constructing the failed drive)

 

Joe L.

Thanks Joe,

 

I'm rebuilding the disk right now using a "warm spare" I had in the case already.

 

The drive did not respond to any smartctl commands after it went down.

 

I have had trouble with this drive before, so I think I will RMA it while I can.

 

Thanks again for the help

 

Cheers,

Kent