6.0-beta14b Dead Disk

[mazma]

Hey All,

 

Running 6.0 Beta 14b with Dynamix webGui v2015.02.25.

 

Haven't looked at the server for a while but just checked in and it was showing I had a faulty disk. For the life of me I couldn't figure out how to run a parity check with this new gui, so I stopped the array, checked the box for maintenance mode. Couldn't find anything for parity check so restarted the array. That disk is now showing as an unassigned device. If I select that disk and re-start the array it shows as "not installed" on the Array Devices tab.

 

Whats the best way forward here? Excerpt from log below.

 

Mar 30 22:46:02 Tower kernel: sd 11:0:0:0: [sdj] CDB:

Mar 30 22:46:02 Tower kernel: cdb[0]=0x28: 28 00 e8 e0 87 bf 00 00 01 00

Mar 30 22:46:02 Tower kernel: blk_update_request: I/O error, dev sdj, sector 3907028927

Mar 30 22:46:02 Tower kernel: buffer_io_error: 1928 callbacks suppressed

Mar 30 22:46:02 Tower kernel: Buffer I/O error on dev sdj1, logical block 3907028864, async page read

Mar 30 22:46:02 Tower kernel: sd 11:0:0:0: [sdj] 

Mar 30 22:46:02 Tower kernel: Result: hostbyte=0x04 driverbyte=0x00

Mar 30 22:46:02 Tower kernel: sd 11:0:0:0: [sdj] CDB:

Mar 30 22:46:02 Tower kernel: cdb[0]=0x28: 28 00 e8 e0 87 c0 00 00 01 00

Mar 30 22:46:02 Tower kernel: blk_update_request: I/O error, dev sdj, sector 3907028928

Mar 30 22:46:02 Tower kernel: Buffer I/O error on dev sdj1, logical block 3907028865, async page read

Mar 30 22:46:02 Tower kernel: sd 11:0:0:0: [sdj] 

Mar 30 22:46:02 Tower kernel: Result: hostbyte=0x04 driverbyte=0x00

Mar 30 22:46:02 Tower kernel: sd 11:0:0:0: [sdj] CDB:

Mar 30 22:46:02 Tower kernel: cdb[0]=0x28: 28 00 e8 e0 87 c1 00 00 06 00

Mar 30 22:46:02 Tower kernel: blk_update_request: I/O error, dev sdj, sector 3907028929

Mar 30 22:46:02 Tower kernel: Buffer I/O error on dev sdj1, logical block 3907028866, async page read

Mar 30 22:46:02 Tower kernel: Buffer I/O error on dev sdj1, logical block 3907028867, async page read

Mar 30 22:46:02 Tower kernel: Buffer I/O error on dev sdj1, logical block 3907028868, async page read

Mar 30 22:46:02 Tower kernel: Buffer I/O error on dev sdj1, logical block 3907028869, async page read

Mar 30 22:46:02 Tower kernel: Buffer I/O error on dev sdj1, logical block 3907028870, async page read

Mar 30 22:46:02 Tower kernel: Buffer I/O error on dev sdj1, logical block 3907028871, async page read

Mar 30 22:46:02 Tower kernel: sd 11:0:0:0: [sdj] 

 

Thanks!

[itimpi]

You cannot run a parity check if any disk is reported as faulty!  As for where you run one from, it is the Main tab under the Array Operation sub-tab (which seems the logical place).  I am not sure if the GUI is clever enough to not offer this option if a disk is faulty.

 

Based on what you have said that drive may well be dead/dying.    Do you have a replacement drive that you can rebuild its contents onto?

 

If you want more diagnostics then a full syslog and also smartctl report (if you can get one) for the problem drive would help.

[mazma]

Of course! Sorry, in all the years I've had unRaid I've never had to replace a disk and with this new guy I was a bit confused.

 

Think I've got it from here. Thanks!