Syslog errors while preclearing a drive with AOC-MV8-SASLP

Seems I've got an issue based on the syslog info below.  I recently installed a Supermicro AOC-MV8-SASLP and one new drive that is preclearing now.  However, should I terminate the preclear based on the errors below?  I think they happened exactly when I executed the preclear script but I'm not 100% sure.

 

Jan 29 21:40:01 Tower kernel: WARNING: at drivers/ata/libata-core.c:5186 ata_qc_issue+0x10b/0x308() (Minor Issues)
Jan 29 21:40:01 Tower kernel: Hardware name: P5Q Premium
Jan 29 21:40:01 Tower kernel: Modules linked in: md_mod xor i2c_i801 i2c_core ahci pata_marvell mvsas libsas scst scsi_transport_sas sky2 (Drive related)
Jan 29 21:40:01 Tower kernel: Pid: 26934, comm: hdparm Not tainted 2.6.32.9-unRAID #8 (Errors)
Jan 29 21:40:01 Tower kernel: Call Trace: (Errors)
Jan 29 21:40:01 Tower kernel:  [<c102449e>] warn_slowpath_common+0x60/0x77 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c10244c2>] warn_slowpath_null+0xd/0x10 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11b624d>] ata_qc_issue+0x10b/0x308 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11ba260>] ata_scsi_translate+0xd1/0xff (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11a816c>] ? scsi_done+0x0/0xd (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11a816c>] ? scsi_done+0x0/0xd (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11baa40>] ata_sas_queuecmd+0x120/0x1d7 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11bc6df>] ? ata_scsi_pass_thru+0x0/0x21d (Errors)
Jan 29 21:40:01 Tower kernel:  [<f844d69a>] sas_queuecommand+0x65/0x20d [libsas] (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11a816c>] ? scsi_done+0x0/0xd (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11a82c0>] scsi_dispatch_cmd+0x147/0x181 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11ace4d>] scsi_request_fn+0x351/0x376 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c1126798>] __blk_run_queue+0x78/0x10c (Errors)
Jan 29 21:40:01 Tower kernel:  [<c1124446>] elv_insert+0x67/0x153 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11245b8>] __elv_add_request+0x86/0x8b (Errors)
Jan 29 21:40:01 Tower kernel:  [<c1129343>] blk_execute_rq_nowait+0x4f/0x73 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11293dc>] blk_execute_rq+0x75/0x91 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11292cc>] ? blk_end_sync_rq+0x0/0x28 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c112636f>] ? get_request+0x204/0x28d (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11269d6>] ? get_request_wait+0x2b/0xd9 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c112c2bf>] sg_io+0x22d/0x30a (Errors)
Jan 29 21:40:01 Tower kernel:  [<c112c5a8>] scsi_cmd_ioctl+0x20c/0x3bc (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11b3257>] sd_ioctl+0x6a/0x8c (Errors)
Jan 29 21:40:01 Tower kernel:  [<c112a420>] __blkdev_driver_ioctl+0x50/0x62 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c112ad1c>] blkdev_ioctl+0x8b0/0x8dc (Errors)
Jan 29 21:40:01 Tower kernel:  [<c1131e2d>] ? kobject_get+0x12/0x17 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c112b0f8>] ? get_disk+0x4a/0x61 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c101b028>] ? kmap_atomic+0x14/0x16 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11334a5>] ? radix_tree_lookup_slot+0xd/0xf (Errors)
Jan 29 21:40:01 Tower kernel:  [<c104a179>] ? filemap_fault+0xb8/0x305 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c1048c43>] ? unlock_page+0x18/0x1b (Errors)
Jan 29 21:40:01 Tower kernel:  [<c1057c63>] ? __do_fault+0x3a7/0x3da (Errors)
Jan 29 21:40:01 Tower kernel:  [<c105985f>] ? handle_mm_fault+0x42d/0x8f1 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c108b6c6>] block_ioctl+0x2a/0x32 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c108b69c>] ? block_ioctl+0x0/0x32 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c10769d5>] vfs_ioctl+0x22/0x67 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c1076f33>] do_vfs_ioctl+0x478/0x4ac (Errors)
Jan 29 21:40:01 Tower kernel:  [<c105dcdd>] ? do_mmap_pgoff+0x232/0x294 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c1076f93>] sys_ioctl+0x2c/0x45 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c1002935>] syscall_call+0x7/0xb (Errors)
Jan 29 21:40:01 Tower kernel: ---[ end trace a21445ad823f10b5 ]---

Seems I've got an issue based on the syslog info below.  I recently installed a Supermicro AOC-MV8-SASLP and one new drive that is preclearing now.  However, should I terminate the preclear based on the errors below?  I think they happened exactly when I executed the preclear script but I'm not 100% sure.

 

Jan 29 21:40:01 Tower kernel: WARNING: at drivers/ata/libata-core.c:5186 ata_qc_issue+0x10b/0x308() (Minor Issues)
Jan 29 21:40:01 Tower kernel: Hardware name: P5Q Premium
Jan 29 21:40:01 Tower kernel: Modules linked in: md_mod xor i2c_i801 i2c_core ahci pata_marvell mvsas libsas scst scsi_transport_sas sky2 (Drive related)
Jan 29 21:40:01 Tower kernel: Pid: 26934, comm: hdparm Not tainted 2.6.32.9-unRAID #8 (Errors)
Jan 29 21:40:01 Tower kernel: Call Trace: (Errors)
Jan 29 21:40:01 Tower kernel:  [<c102449e>] warn_slowpath_common+0x60/0x77 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c10244c2>] warn_slowpath_null+0xd/0x10 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11b624d>] ata_qc_issue+0x10b/0x308 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11ba260>] ata_scsi_translate+0xd1/0xff (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11a816c>] ? scsi_done+0x0/0xd (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11a816c>] ? scsi_done+0x0/0xd (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11baa40>] ata_sas_queuecmd+0x120/0x1d7 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11bc6df>] ? ata_scsi_pass_thru+0x0/0x21d (Errors)
Jan 29 21:40:01 Tower kernel:  [<f844d69a>] sas_queuecommand+0x65/0x20d [libsas] (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11a816c>] ? scsi_done+0x0/0xd (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11a82c0>] scsi_dispatch_cmd+0x147/0x181 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11ace4d>] scsi_request_fn+0x351/0x376 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c1126798>] __blk_run_queue+0x78/0x10c (Errors)
Jan 29 21:40:01 Tower kernel:  [<c1124446>] elv_insert+0x67/0x153 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11245b8>] __elv_add_request+0x86/0x8b (Errors)
Jan 29 21:40:01 Tower kernel:  [<c1129343>] blk_execute_rq_nowait+0x4f/0x73 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11293dc>] blk_execute_rq+0x75/0x91 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11292cc>] ? blk_end_sync_rq+0x0/0x28 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c112636f>] ? get_request+0x204/0x28d (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11269d6>] ? get_request_wait+0x2b/0xd9 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c112c2bf>] sg_io+0x22d/0x30a (Errors)
Jan 29 21:40:01 Tower kernel:  [<c112c5a8>] scsi_cmd_ioctl+0x20c/0x3bc (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11b3257>] sd_ioctl+0x6a/0x8c (Errors)
Jan 29 21:40:01 Tower kernel:  [<c112a420>] __blkdev_driver_ioctl+0x50/0x62 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c112ad1c>] blkdev_ioctl+0x8b0/0x8dc (Errors)
Jan 29 21:40:01 Tower kernel:  [<c1131e2d>] ? kobject_get+0x12/0x17 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c112b0f8>] ? get_disk+0x4a/0x61 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c101b028>] ? kmap_atomic+0x14/0x16 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c11334a5>] ? radix_tree_lookup_slot+0xd/0xf (Errors)
Jan 29 21:40:01 Tower kernel:  [<c104a179>] ? filemap_fault+0xb8/0x305 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c1048c43>] ? unlock_page+0x18/0x1b (Errors)
Jan 29 21:40:01 Tower kernel:  [<c1057c63>] ? __do_fault+0x3a7/0x3da (Errors)
Jan 29 21:40:01 Tower kernel:  [<c105985f>] ? handle_mm_fault+0x42d/0x8f1 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c108b6c6>] block_ioctl+0x2a/0x32 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c108b69c>] ? block_ioctl+0x0/0x32 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c10769d5>] vfs_ioctl+0x22/0x67 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c1076f33>] do_vfs_ioctl+0x478/0x4ac (Errors)
Jan 29 21:40:01 Tower kernel:  [<c105dcdd>] ? do_mmap_pgoff+0x232/0x294 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c1076f93>] sys_ioctl+0x2c/0x45 (Errors)
Jan 29 21:40:01 Tower kernel:  [<c1002935>] syscall_call+0x7/0xb (Errors)
Jan 29 21:40:01 Tower kernel: ---[ end trace a21445ad823f10b5 ]---

It was the hdparm commad, and since the preclear script does not use hdparm, odds are good it has nothing to do with the preclear itself directly.  It does look like "hdparm" is trying to do a memory mapping, so perhaps you were running short of free memory at the time the command was invoked.

What exactly does the HDPARM command trying to do a memory mapping mean and is this something that I invoked (albeit inadvertently)?  Just wondering if this is a random error and I shouldn't panic or if I should do anything?  Preclear of the new 2TB drive has been running about 11.5 hrs and I'm 77% through step 2.  It's writing at about 95 MB/sec atm.

Looks like I am having a similar type of issue, also while preclearing a drive with AOC-MV8-SASLP.  I am running unRAID 4.7.  Was not aware of the issue at the time and preclearing the disk nearly finished with this output:

================================================================== 1.3
=                unRAID server Pre-Clear disk /dev/sdb
=               cycle 1 of 1, partition start on sector 63 
= Disk Pre-Clear-Read completed                                 DONE
= Step 1 of 10 - Copying zeros to first 2048k bytes             DONE
= Step 2 of 10 - Copying zeros to remainder of disk to clear it DONE
= Step 3 of 10 - Disk is now cleared from MBR onward.           DONE
= Step 4 of 10 - Clearing MBR bytes for partition 2,3 & 4       DONE
= Step 5 of 10 - Clearing MBR code area                         DONE
= Step 6 of 10 - Setting MBR signature bytes                    DONE
= Step 7 of 10 - Setting partition 1 to precleared state        DONE
= Step 8 of 10 - Notifying kernel we changed the partitioning   DONE
= Step 10 of 10 - Verifying if the MBR is cleared.     DONE
= Step 10 of 10 - Verifying the clear has been successful.
= 
Elapsed Time:  5:09:03
========================================================================1.3
==
== SORRY: Disk /dev/sdb MBR could NOT be precleared
==
== out4= 00000
== out5= 00000
============================================================================
dd: reading `/dev/sdb': Input/output error
0+0 records in
0+0 records out
0 bytes (0 B) copied, 0.00160582 s, 0.0 kB/s
0000000

 

My syslog grew to 2 gigs.  Looks like I got a similar message as betaman just after starting the preclear.

Feb  2 04:46:24 Tower kernel: WARNING: at drivers/ata/libata-core.c:5186 ata_qc_issue+0x10b/0x308()
Feb  2 04:46:24 Tower kernel: Hardware name: A760G M2+
Feb  2 04:46:24 Tower kernel: Modules linked in: md_mod xor atiixp ahci r8169 mvsas libsas scst scsi_transport_sas
Feb  2 04:46:24 Tower kernel: Pid: 2047, comm: smartctl Not tainted 2.6.32.9-unRAID #8
Feb  2 04:46:24 Tower kernel: Call Trace:
Feb  2 04:46:24 Tower kernel:  [<c102449e>] warn_slowpath_common+0x60/0x77
Feb  2 04:46:24 Tower kernel:  [<c10244c2>] warn_slowpath_null+0xd/0x10
Feb  2 04:46:24 Tower kernel:  [<c11b624d>] ata_qc_issue+0x10b/0x308
Feb  2 04:46:24 Tower kernel:  [<c11ac0ba>] ? scsi_init_sgtable+0x5b/0x76
Feb  2 04:46:24 Tower kernel:  [<c11ba260>] ata_scsi_translate+0xd1/0xff
Feb  2 04:46:24 Tower kernel:  [<c11a816c>] ? scsi_done+0x0/0xd
Feb  2 04:46:24 Tower kernel:  [<c11a816c>] ? scsi_done+0x0/0xd
Feb  2 04:46:24 Tower kernel:  [<c11baa40>] ata_sas_queuecmd+0x120/0x1d7
Feb  2 04:46:24 Tower kernel:  [<c11bc6df>] ? ata_scsi_pass_thru+0x0/0x21d
Feb  2 04:46:24 Tower kernel:  [<f843769a>] sas_queuecommand+0x65/0x20d [libsas]
Feb  2 04:46:24 Tower kernel:  [<c11a816c>] ? scsi_done+0x0/0xd
Feb  2 04:46:24 Tower kernel:  [<c11a82c0>] scsi_dispatch_cmd+0x147/0x181
Feb  2 04:46:24 Tower kernel:  [<c11ace4d>] scsi_request_fn+0x351/0x376
Feb  2 04:46:24 Tower kernel:  [<c1126798>] __blk_run_queue+0x78/0x10c
Feb  2 04:46:24 Tower kernel:  [<c1124446>] elv_insert+0x67/0x153
Feb  2 04:46:24 Tower kernel:  [<c11245b8>] __elv_add_request+0x86/0x8b
Feb  2 04:46:24 Tower kernel:  [<c1129343>] blk_execute_rq_nowait+0x4f/0x73
Feb  2 04:46:24 Tower kernel:  [<c11293dc>] blk_execute_rq+0x75/0x91
Feb  2 04:46:24 Tower kernel:  [<c11292cc>] ? blk_end_sync_rq+0x0/0x28
Feb  2 04:46:24 Tower kernel:  [<c11295fa>] ? blk_recount_segments+0x16/0x24
Feb  2 04:46:24 Tower kernel:  [<c11250c6>] ? blk_rq_bio_prep+0x47/0x78
Feb  2 04:46:24 Tower kernel:  [<c1128edb>] ? blk_rq_append_bio+0x14/0x3b
Feb  2 04:46:24 Tower kernel:  [<c1129121>] ? blk_rq_map_user+0x12c/0x1be
Feb  2 04:46:24 Tower kernel:  [<c112c2bf>] sg_io+0x22d/0x30a
Feb  2 04:46:24 Tower kernel:  [<c112c5a8>] scsi_cmd_ioctl+0x20c/0x3bc
Feb  2 04:46:24 Tower kernel:  [<c104cbfe>] ? get_page_from_freelist+0x2e4/0x344
Feb  2 04:46:24 Tower kernel:  [<c11b3257>] sd_ioctl+0x6a/0x8c
Feb  2 04:46:24 Tower kernel:  [<c112a420>] __blkdev_driver_ioctl+0x50/0x62
Feb  2 04:46:24 Tower kernel:  [<c112ad1c>] blkdev_ioctl+0x8b0/0x8dc
Feb  2 04:46:24 Tower kernel:  [<c1131e2d>] ? kobject_get+0x12/0x17
Feb  2 04:46:24 Tower kernel:  [<c112b0f8>] ? get_disk+0x4a/0x61
Feb  2 04:46:24 Tower kernel:  [<c101b028>] ? kmap_atomic+0x14/0x16
Feb  2 04:46:24 Tower kernel:  [<c11334a5>] ? radix_tree_lookup_slot+0xd/0xf
Feb  2 04:46:24 Tower kernel:  [<c104a179>] ? filemap_fault+0xb8/0x305
Feb  2 04:46:24 Tower kernel:  [<c1048c43>] ? unlock_page+0x18/0x1b
Feb  2 04:46:24 Tower kernel:  [<c1057c63>] ? __do_fault+0x3a7/0x3da
Feb  2 04:46:24 Tower kernel:  [<c105985f>] ? handle_mm_fault+0x42d/0x8f1
Feb  2 04:46:24 Tower kernel:  [<c108b6c6>] block_ioctl+0x2a/0x32
Feb  2 04:46:24 Tower kernel:  [<c108b69c>] ? block_ioctl+0x0/0x32
Feb  2 04:46:24 Tower kernel:  [<c10769d5>] vfs_ioctl+0x22/0x67
Feb  2 04:46:24 Tower kernel:  [<c1076f33>] do_vfs_ioctl+0x478/0x4ac
Feb  2 04:46:24 Tower kernel:  [<c105c983>] ? expand_downwards+0x109/0x136
Feb  2 04:46:24 Tower kernel:  [<c1076f93>] sys_ioctl+0x2c/0x45
Feb  2 04:46:24 Tower kernel:  [<c1002935>] syscall_call+0x7/0xb
Feb  2 04:46:24 Tower kernel: ---[ end trace cb967b7456a0ba22 ]---

 

Any help would be appreciated.

 

Thanks

 

 

I seem to have run into this recently as well... same symptoms..

 

syslog file growing very large sometimes to 3+gigs.

 

Same errors in the syslog...

 

unRaid 4.7 as well with the AOC-SASLP controller... I think I might have isolated it to a bad drive. I bought 4 new drives and re-used 2 good drives. 1 of the new drives (Hitachi 5k3000 2TB) seem to be bad, freezing preclears and causing these issues. I've removed that drive and so far pre-clearing all the other drives ok.

 

 

I've had this happen as well.  I took (3) previously precleard 20EARS and tried to preclear them on the AOC-MV8-SASLP and one precleared fine but really slow and the other two kept locking up while preclearing.  I then precleared those same drives on the motherboard sata ports without problems.  I haven't trusted it yet enought to put my array drives on it yet.

Did anyone come to a conclusion on this AOC-MV8-SASLP card?

 

Unfortunately I have the same card and can't pre-clear on it... not sure I want to put the card into use if that is the case.

I remember reading somewhere that this happens when you add a second AOC-MV8-SASLP to your setup.  I have had two cards in my array since day one.  I have the same errors in my syslog and Parity checks are still fine and I have replaced 7 old drives with newer larger drives in my array.  When I first started I had issues with connections of power and sata drive.  After fixing those all has been working fine.  The AOC-MV8-SASLP is rock solid in my setup.

Thanks for the reply.

 

I am wondering if there is something to the cabling/power in my situation.  The only problem is that i used the same power connectors on on the drives between moving them from the onboard SATA ports and the AOC-MV8. The SATA cables are the breakout type, with 4 coming off each condensed connection on the card, you know what I mean there.  I could swap those for one another (I have two sets of 4) and try I suppose.

 

I have the Corsair PSU listed in my sig, maybe that is under powered?  Though, I haven't ran more than 3 drives at a time in this thing...

It's pretty bad when the OP can't remember exactly what ended up happening!!  Anyway, I can tell you that I successfully pre-cleared two drives with this card.  I don't think I got the errors on the second drive but the first drive completed without restarting.  I've been using the card now for several months without issue.

Hi all is there a solution for this issue? im having the same issue im Preclearing a WD2EARS on the mboard (120MB/sec) at the same time no problems

the WD1.5EARS (2.5MB/sec) is on the AOC-MV8-SASLP

 

Aug  1 19:54:28 Tower kernel: Pid: 3259, comm: hdparm Not tainted 2.6.32.9-unRAID #8 (Errors)
Aug  1 19:54:28 Tower kernel: Call Trace: (Errors)
Aug  1 19:54:28 Tower kernel:  [<c102449e>] warn_slowpath_common+0x60/0x77 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c10244c2>] warn_slowpath_null+0xd/0x10 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c11b624d>] ata_qc_issue+0x10b/0x308 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c11ba260>] ata_scsi_translate+0xd1/0xff (Errors)
Aug  1 19:54:28 Tower kernel:  [<c11a816c>] ? scsi_done+0x0/0xd (Errors)
Aug  1 19:54:28 Tower kernel:  [<c11a816c>] ? scsi_done+0x0/0xd (Errors)
Aug  1 19:54:28 Tower kernel:  [<c11baa40>] ata_sas_queuecmd+0x120/0x1d7 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c11bc6df>] ? ata_scsi_pass_thru+0x0/0x21d (Errors)
Aug  1 19:54:28 Tower kernel:  [<f842969a>] sas_queuecommand+0x65/0x20d [libsas] (Errors)
Aug  1 19:54:28 Tower kernel:  [<c11a816c>] ? scsi_done+0x0/0xd (Errors)
Aug  1 19:54:28 Tower kernel:  [<c11a82c0>] scsi_dispatch_cmd+0x147/0x181 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c11ace4d>] scsi_request_fn+0x351/0x376 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c1126798>] __blk_run_queue+0x78/0x10c (Errors)
Aug  1 19:54:28 Tower kernel:  [<c1124446>] elv_insert+0x67/0x153 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c11245b8>] __elv_add_request+0x86/0x8b (Errors)
Aug  1 19:54:28 Tower kernel:  [<c1129343>] blk_execute_rq_nowait+0x4f/0x73 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c11293dc>] blk_execute_rq+0x75/0x91 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c11292cc>] ? blk_end_sync_rq+0x0/0x28 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c112636f>] ? get_request+0x204/0x28d (Errors)
Aug  1 19:54:28 Tower kernel:  [<c11269d6>] ? get_request_wait+0x2b/0xd9 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c112c2bf>] sg_io+0x22d/0x30a (Errors)
Aug  1 19:54:28 Tower kernel:  [<c112c5a8>] scsi_cmd_ioctl+0x20c/0x3bc (Errors)
Aug  1 19:54:28 Tower kernel:  [<c11b3257>] sd_ioctl+0x6a/0x8c (Errors)
Aug  1 19:54:28 Tower kernel:  [<c112a420>] __blkdev_driver_ioctl+0x50/0x62 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c112ad1c>] blkdev_ioctl+0x8b0/0x8dc (Errors)
Aug  1 19:54:28 Tower kernel:  [<c1131e2d>] ? kobject_get+0x12/0x17 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c112b0f8>] ? get_disk+0x4a/0x61 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c101b028>] ? kmap_atomic+0x14/0x16 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c11334a5>] ? radix_tree_lookup_slot+0xd/0xf (Errors)
Aug  1 19:54:28 Tower kernel:  [<c104a179>] ? filemap_fault+0xb8/0x305 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c1048c43>] ? unlock_page+0x18/0x1b (Errors)
Aug  1 19:54:28 Tower kernel:  [<c1057c63>] ? __do_fault+0x3a7/0x3da (Errors)
Aug  1 19:54:28 Tower kernel:  [<c105985f>] ? handle_mm_fault+0x42d/0x8f1 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c108b6c6>] block_ioctl+0x2a/0x32 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c108b69c>] ? block_ioctl+0x0/0x32 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c10769d5>] vfs_ioctl+0x22/0x67 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c1076f33>] do_vfs_ioctl+0x478/0x4ac (Errors)
Aug  1 19:54:28 Tower kernel:  [<c105dcdd>] ? do_mmap_pgoff+0x232/0x294 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c1076f93>] sys_ioctl+0x2c/0x45 (Errors)
Aug  1 19:54:28 Tower kernel:  [<c1002935>] syscall_call+0x7/0xb (Errors)

 

any advice would be great as ive only just lost alot of data due to double disk fails

what board are you using?  What BIOS on the AOC card?

 

I am pre-clearing all my drives on the onboard now - when that is done I will mess with my AOC card a bit, try upgrading the BIOS, etc.  Perhaps you can try different breakout SATA cables?  I have two sets, was going to try using the set for drives 5-8 and see if that is better.

 

There are a lot of suggestions to replace SATA cables, etc but I can't believe in this day and age those are still issues. We get Dells by the 1000's and get a DOA or problematic machine about 1/500... and it's never the SATA cables, I can tell you that much.

Im ashamed to say im not actually sure what BIOS im using as i purchased the card and just slapped it in

looking back at the preclear session it seems to have picked up again (80MB /sec)

hopefully it will pan out

on the note of SATA cables i work for in technical for DVR company and we import and build a few differnt DVRs and we must go through 1000's of drives a month (WD blue, black, green depending on the DVR) and i can tell you we have never had one back with SATA issues although the fail rate on WD20EARS is suprisingly high

It's interesting that it picked back up... and again, this is preclearing on the AOC and not on the onboard?  I have mine rigged to send me emails to my gmail account along each step of the way. When I was pre-clearing on the AOC card I would get the notice, then basically nothing - let it go a day or more.  UnRAID was still going but the pre-clear sessions seemed frozen (didn't look at the unRAID syslog, foolishly).

 

On the onboard, those same two drives were on post-read at 108mb/s each (both at same time) based on the last email I got.

 

Are these DVR's for security systems by chance?  I got some piecemeal system (something like 'Homeland Security' brand) used a year ago to record by my garage out back and I am unable to login remotely via the webinterface, but the keyboard/mouse/monitor hookups work - wonder if there is a way to reset the PW. Can't even find the manual online anywhere - it seems to be an off, off brand  ... I have seen pics online of other systems that have the same case, etc, just branded different.