grover11

Yes it was excluded after the share was created, it was a disk added after the share was created.

I have shares copying to excluded drives, they are set to copy to most free but periodically copy to excluded drives. Any ideas why?

I opened the case and reseated all the cables. It rebuilt the parity and looks good so far. I'm thinking if I'm gonna have issues they will arise when I write to the array not when I am just reading from it. I'll write something to it later today.

Could this be part of the problem? parity device: pci-0000:00:1f.2-scsi-0:0:0:0 host1 (sda) ST2000DL003-9VT166_5YD2Z590 disk1 device: pci-0000:00:1f.5-scsi-1:0:0:0 host4 (sdf) ST2000DL003-9VT166_5YD2TKNM disk2 device: pci-0000:00:1f.2-scsi-1:0:1:0 host2 (sdd) ST2000DL003-9VT166_5YD2Y1K5 disk3 device: pci-0000:00:1f.5-scsi-0:0:0:0 host3 (sde) ST2000DL003-9VT166_5YD2VWT7 disk4 device: pci-0000:00:1f.2-scsi-0:0:1:0 host1 (sdb) ST2000DL003-9VT166_5YD28YGC disk5 device: pci-0000:00:1f.2-scsi-1:0:0:0 host2 (sdc) ST2000DL003-9VT166_5YD31EBL It's showing 2 disks on Host 1 and 2 disks on Host 2

I am back where I started, Parity disk showing as new and rebuilding again. Appears to be having issues with disk 4 as well. Sep 3 09:49:35 Tower kernel: handle_stripe write error: 2192388096/4, count: 1 (Errors) Sep 3 09:49:35 Tower kernel: md: disk4 write error (Errors) Sep 3 09:49:35 Tower kernel: handle_stripe write error: 2192388104/4, count: 1 (Errors) Sep 3 09:49:35 Tower kernel: md: disk4 write error (Errors) Sep 3 09:49:35 Tower kernel: handle_stripe write error: 2192388112/4, count: 1 (Errors) Sep 3 09:49:35 Tower kernel: md: disk4 write error (Errors) Sep 3 09:49:35 Tower kernel: handle_stripe write error: 2192388120/4, count: 1 (Errors) Sep 3 09:49:35 Tower kernel: md: disk4 write error (Errors) Sep 3 09:49:35 Tower kernel: handle_stripe write error: 2192388128/4, count: 1 (Errors) Sep 3 09:49:35 Tower kernel: md: disk4 write error (Errors) Sep 3 09:49:35 Tower kernel: handle_stripe write error: 2192388136/4, count: 1 (Errors) Sep 3 09:49:35 Tower kernel: md: disk4 write error (Errors) Sep 3 09:49:35 Tower kernel: handle_stripe write error: 2192388144/4, count: 1 (Errors) Sep 3 09:49:35 Tower kernel: md: disk4 write error (Errors) Sep 3 09:49:35 Tower kernel: handle_stripe write error: 2192388152/4, count: 1 (Errors)

I stopped the array, shutdown the serger and re-seated the drive. It restarted without any errors and I was able to do the short smart test, it showed no errors. Hopefully that was the problem.

Now I get this error when trying to run the short smart test from unmenu smartctl -t short -d ata /dev/sda 2>&1 smartctl 5.39.1 2010-01-28 r3054 [i486-slackware-linux-gnu] (local build) Copyright © 2002-10 by Bruce Allen, http://smartmontools.sourceforge.net Smartctl: Device Read Identity Failed (not an ATA/ATAPI device) A mandatory SMART command failed: exiting. To continue, add one or more '-T permissive' options. HELP!

Actually it looks like I may have a problem with the parity disk. Sep 3 09:07:53 Tower kernel: handle_stripe read error: 1346904/0, count: 1 (Errors) Sep 3 09:07:53 Tower kernel: md: disk0 read error (Errors) Sep 3 09:07:53 Tower kernel: handle_stripe read error: 1346912/0, count: 1 (Errors) Sep 3 09:07:53 Tower kernel: md: disk0 read error (Errors) Sep 3 09:07:53 Tower kernel: handle_stripe read error: 1346920/0, count: 1 (Errors) Sep 3 09:07:53 Tower kernel: md: disk0 read error (Errors) Sep 3 09:07:53 Tower kernel: handle_stripe read error: 1346928/0, count: 1 (Errors) Sep 3 09:07:53 Tower kernel: md: disk0 read error (Errors) Sep 3 09:07:53 Tower kernel: handle_stripe read error: 1346936/0, count: 1 (Errors) Sep 3 09:07:53 Tower kernel: md: disk0 read error (Errors) Sep 3 09:07:53 Tower kernel: handle_stripe read error: 1346944/0, count: 1 (Errors) Sep 3 09:07:54 Tower kernel: program smartctl is using a deprecated SCSI ioctl, please convert it to SG_IO Sep 3 09:07:54 Tower last message repeated 3 times Sep 3 09:07:54 Tower kernel: NTFS driver 2.1.29 [Flags: R/O MODULE]. (System) Sep 3 09:08:34 Tower kernel: program smartctl is using a deprecated SCSI ioctl, please convert it to SG_IO

I let the parity re-synch(completed with no errors) shut the server down for the night and when I started this morning all was well.

I'm running identical hardware to the 20 drive Tower from Raj. I haven't added the controller cards yet so I have all 6 drives connected to the motherboard. The drives are Seagate Barracuda Greens 2TB, all were successfully precleared before being added to the array. I rebooted the server after I got the error and the syslog reset and only has entries since after the reboot so I don't think that will be much help. It is rebuilding the parity on the same original parity disk right now. I just wonder if moving and deleting folders from one share to another as well as reassigning the disks to the shares might have screwed things up.

I have had UnRaid 4.7 running for a few days. I added the parity disk a couple of days ago and today (after moving one folder from one disk to another )it showed as a new disk and is rebuilding parity again after a reboot of the server. Any clues as to why?

Thanks Alot, feel a little dumb but thats ok.

I can't see a link to get the preclear script, am I missing something?

OK, they wanted $56 to ship 3 cables to Canada so I found them at a Canadian site for $19 each and $7 total shipping. Thanks again for your help. Don't be surprised if I'm on here looking for more guidance later.

Johnm Thanks for the reply, just to be clear I need 4 of these, 3ware CBL-SFF8087OCF-05M Multi-lane Internal (SFF-8087) Serial ATA breakout cable ? The cables are the last thing I have to sort out and get ordered. I ordered one of these already so hopefully they are the right ones and I will get 3 more.