Message-ID: <20071024205212.GO4003@csclub.uwaterloo.ca>
Date:	Wed, 24 Oct 2007 16:52:12 -0400
From:	lsorense@...lub.uwaterloo.ca (Lennart Sorensen)
To:	"Nestor A. Diaz" <nestor@...ndalinux.com>
Cc:	linux-kernel@...r.kernel.org
Subject: Re: RAID 10 w AHCI w NCQ = Spurius  I/O error

On Wed, Oct 24, 2007 at 03:17:27PM -0500, Nestor A. Diaz wrote:
> Hello People,
> 
> I need your help, this problem is turning me crazy.
> 
> I have created a RAID 10 using  a RAID0 configuration on top of a two 
> RAID1 devices (all software raid), like this:
> 
> Personalities : [raid0] [raid1]
> md4 : active raid0 md2[0] md3[1]
>      605071872 blocks 64k chunks
> 
> md0 : active raid1 sdd3[3] sda3[0] sdc3[2] sdb3[1]
>      9791552 blocks [4/4] [UUUU]
> 
> md3 : active raid1 sdd2[2](F) sdb2[0]
>      302536000 blocks [2/1] [U_]
> 
> md1 : active raid1 sdd1[3] sda1[0] sdc1[2] sdb1[1]
>      240832 blocks [4/4] [UUUU]
> 
> md2 : active raid1 sda2[0] sdc2[1]
>      302536000 blocks [2/2] [UU]
> 
> unused devices: <none>
> 
> But the sdd device sometimes fail, i have changed the hard disk, check 
> the older sata drive, reformat using mke2fs -c -c (to check for media 
> errrors both read and write, no media problems found, change the sata 
> disk and the problem remains, also with a new sata hard disk).
> 
> The systema is a supermicro server 5015-mt+ with an ich7 ahci controller
> 
> All SATA hard disk are hitachi, here is a copy of the kernel log:
> 
> Oct 19 14:01:08 rogue kernel: scsi0 : ahci
> Oct 19 14:01:08 rogue kernel: scsi1 : ahci
> Oct 19 14:01:08 rogue kernel: scsi2 : ahci
> Oct 19 14:01:08 rogue kernel: scsi3 : ahci
> Oct 19 14:01:08 rogue kernel: ata1: SATA max UDMA/133 cmd 
> 0xffffc20001434500 ctl 0x0000000000000000 bmdma 0x00000000
> 00000000 irq 1275
> Oct 19 14:01:08 rogue kernel: ata2: SATA max UDMA/133 cmd 
> 0xffffc20001434580 ctl 0x0000000000000000 bmdma 0x00000000
> 00000000 irq 1275
> Oct 19 14:01:08 rogue kernel: ata3: SATA max UDMA/133 cmd 
> 0xffffc20001434600 ctl 0x0000000000000000 bmdma 0x00000000
> 00000000 irq 1275
> Oct 19 14:01:08 rogue kernel: ata4: SATA max UDMA/133 cmd 
> 0xffffc20001434680 ctl 0x0000000000000000 bmdma 0x00000000
> 00000000 irq 1275
> Oct 19 14:01:08 rogue kernel: ata1: SATA link up 3.0 Gbps (SStatus 123 
> SControl 300)
> Oct 19 14:01:08 rogue kernel: ata1.00: ATA-7: Hitachi HDT725032VLA360, 
> V54OA73A, max UDMA/133
> Oct 19 14:01:08 rogue kernel: ata1.00: 625142448 sectors, multi 16: 
> LBA48 NCQ (depth 31/32)
> Oct 19 14:01:08 rogue kernel: ata1.00: configured for UDMA/133
> Oct 19 14:01:08 rogue kernel: ata2: SATA link up 3.0 Gbps (SStatus 123 
> SControl 300)
> Oct 19 14:01:08 rogue kernel: ata2.00: ATA-7: Hitachi HDT725032VLA360, 
> V54OA73A, max UDMA/133
> Oct 19 14:01:08 rogue kernel: ata2.00: 625142448 sectors, multi 16: 
> LBA48 NCQ (depth 31/32)
> Oct 19 14:01:08 rogue kernel: ata2.00: configured for UDMA/133
> Oct 19 14:01:08 rogue kernel: ata3: SATA link up 3.0 Gbps (SStatus 123 
> SControl 300)
> Oct 19 14:01:08 rogue kernel: ata3.00: ATA-7: Hitachi HDT725032VLA360, 
> V54OA73A, max UDMA/133
> Oct 19 14:01:08 rogue kernel: ata3.00: 625142448 sectors, multi 16: 
> LBA48 NCQ (depth 31/32)
> Oct 19 14:01:08 rogue kernel: ata3.00: configured for UDMA/133
> Oct 19 14:01:08 rogue kernel: ata4: SATA link up 3.0 Gbps (SStatus 123 
> SControl 300)
> Oct 19 14:01:08 rogue kernel: ata4.00: ATA-7: Hitachi HDT725032VLA360, 
> V54OA73A, max UDMA/133
> Oct 19 14:01:08 rogue kernel: ata4.00: 625142448 sectors, multi 16: 
> LBA48 NCQ (depth 31/32)
> Oct 19 14:01:08 rogue kernel: ata4.00: configured for UDMA/133
> Oct 19 14:01:08 rogue kernel: scsi 0:0:0:0: Direct-Access     ATA      
> Hitachi HDT72503 V54O PQ: 0 ANSI: 5
> Oct 19 14:01:08 rogue kernel: scsi 1:0:0:0: Direct-Access     ATA      
> Hitachi HDT72503 V54O PQ: 0 ANSI: 5
> Oct 19 14:01:08 rogue kernel: scsi 2:0:0:0: Direct-Access     ATA      
> Hitachi HDT72503 V54O PQ: 0 ANSI: 5
> Oct 19 14:01:08 rogue kernel: scsi 3:0:0:0: Direct-Access     ATA      
> Hitachi HDT72503 V54O PQ: 0 ANSI: 5
> Oct 19 14:01:08 rogue kernel: sd 0:0:0:0: [sda] 625142448 512-byte 
> hardware sectors (320073 MB)
> Oct 19 14:01:08 rogue kernel: sd 0:0:0:0: [sda] Write Protect is off
> Oct 19 14:01:08 rogue kernel: sd 0:0:0:0: [sda] Mode Sense: 00 3a 00 00
> Oct 19 14:01:08 rogue kernel: hda: ATAPI<5>sd 0:0:0:0: [sda] Write 
> cache: enabled, read cache: enabled, doesn't supp
> ort DPO or FUA
> Oct 19 14:01:08 rogue kernel: sd 0:0:0:0: [sda] 625142448 512-byte 
> hardware sectors (320073 MB)
> Oct 19 14:01:08 rogue kernel: sd 0:0:0:0: [sda] Write Protect is off
> Oct 19 14:01:08 rogue kernel: sd 0:0:0:0: [sda] Mode Sense: 00 3a 00 00
> Oct 19 14:01:08 rogue kernel: sd 0:0:0:0: [sda] Write cache: enabled, 
> read cache: enabled, doesn't support DPO or FU
> A
> Oct 19 14:01:08 rogue kernel: sda: sda1 sda2 sda3
> Oct 19 14:01:08 rogue kernel: 24X<5>sd 0:0:0:0: [sda] Attached SCSI disk
> Oct 19 14:01:08 rogue kernel: sd 1:0:0:0: [sdb] 625142448 512-byte 
> hardware sectors (320073 MB)
> Oct 19 14:01:08 rogue kernel: sd 1:0:0:0: [sdb] Write Protect is off
> Oct 19 14:01:08 rogue kernel: sd 1:0:0:0: [sdb] Mode Sense: 00 3a 00 00
> Oct 19 14:01:08 rogue kernel: sd 1:0:0:0: [sdb] Write cache: enabled, 
> read cache: enabled, doesn't support DPO or FU
> A
> Oct 19 14:01:08 rogue kernel: sd 1:0:0:0: [sdb] 625142448 512-byte 
> hardware sectors (320073 MB)
> Oct 19 14:01:08 rogue kernel: sd 1:0:0:0: [sdb] Write Protect is off
> Oct 19 14:01:08 rogue kernel: sd 1:0:0:0: [sdb] Mode Sense: 00 3a 00 00
> Oct 19 14:01:08 rogue kernel: sd 1:0:0:0: [sdb] Write cache: enabled, 
> read cache: enabled, doesn't support DPO or FU
> A
> Oct 19 14:01:08 rogue kernel: sdb: sdb1 sdb2 sdb3
> Oct 19 14:01:08 rogue kernel: CD-ROM<5>sd 1:0:0:0: [sdb] Attached SCSI disk
> Oct 19 14:01:08 rogue kernel: sd 2:0:0:0: [sdc] 625142448 512-byte 
> hardware sectors (320073 MB)
> Oct 19 14:01:08 rogue kernel: sd 2:0:0:0: [sdc] Write Protect is off
> Oct 19 14:01:08 rogue kernel: sd 2:0:0:0: [sdc] Mode Sense: 00 3a 00 00
> Oct 19 14:01:08 rogue kernel: sd 2:0:0:0: [sdc] Write cache: enabled, 
> read cache: enabled, doesn't support DPO or FU
> A
> Oct 19 14:01:08 rogue kernel: sd 2:0:0:0: [sdc] 625142448 512-byte 
> hardware sectors (320073 MB)
> Oct 19 14:01:08 rogue kernel: sd 2:0:0:0: [sdc] Write Protect is off
> Oct 19 14:01:08 rogue kernel: sd 2:0:0:0: [sdc] Mode Sense: 00 3a 00 00
> Oct 19 14:01:08 rogue kernel: sd 2:0:0:0: [sdc] Write cache: enabled, 
> read cache: enabled, doesn't support DPO or FU
> A
> Oct 19 14:01:08 rogue kernel: sdc: sdc1 sdc2 sdc3
> Oct 19 14:01:08 rogue kernel: drive<5>sd 2:0:0:0: [sdc] Attached SCSI disk
> Oct 19 14:01:08 rogue kernel: sd 3:0:0:0: [sdd] 625142448 512-byte 
> hardware sectors (320073 MB)
> Oct 19 14:01:08 rogue kernel: sd 3:0:0:0: [sdd] Write Protect is off
> Oct 19 14:01:08 rogue kernel: sd 3:0:0:0: [sdd] Mode Sense: 00 3a 00 00
> Oct 19 14:01:08 rogue kernel: sd 3:0:0:0: [sdd] Write cache: enabled, 
> read cache: enabled, doesn't support DPO or FU
> A
> Oct 19 14:01:08 rogue kernel: sd 3:0:0:0: [sdd] 625142448 512-byte 
> hardware sectors (320073 MB)
> Oct 19 14:01:08 rogue kernel: sd 3:0:0:0: [sdd] Write Protect is off
> Oct 19 14:01:08 rogue kernel: sd 3:0:0:0: [sdd] Mode Sense: 00 3a 00 00
> Oct 19 14:01:08 rogue kernel: sd 3:0:0:0: [sdd] Write cache: enabled, 
> read cache: enabled, doesn't support DPO or FU
> A
> Oct 19 14:01:08 rogue kernel: Uniform CD-ROM driver Revision: 3.20
> Oct 19 14:01:08 rogue kernel: Probing IDE interface ide1...
> Oct 19 14:01:08 rogue kernel: md: raid0 personality registered for level 0
> Oct 19 14:01:08 rogue kernel: md: raid1 personality registered for level 1
> Oct 19 14:01:08 rogue kernel: md: md2 stopped.
> Oct 19 14:01:08 rogue kernel: md: bind<sdc2>
> Oct 19 14:01:08 rogue kernel: md: bind<sda2>
> Oct 19 14:01:08 rogue kernel: raid1: raid set md2 active with 2 out of 2 
> mirrors
> Oct 19 14:01:08 rogue kernel: md: md1 stopped.
> Oct 19 14:01:08 rogue kernel: md: bind<sdb1>
> Oct 19 14:01:08 rogue kernel: md: bind<sdc1>
> Oct 19 14:01:08 rogue kernel: md: bind<sda1>
> Oct 19 14:01:08 rogue kernel: raid1: raid set md1 active with 3 out of 4 
> mirrors
> Oct 19 14:01:08 rogue kernel: md: md3 stopped.
> Oct 19 14:01:08 rogue kernel: md: bind<sdb2>
> Oct 19 14:01:08 rogue kernel: raid1: raid set md3 active with 1 out of 2 
> mirrors
> Oct 19 14:01:08 rogue kernel: md: md0 stopped.
> Oct 19 14:01:08 rogue kernel: md: bind<sdb3>
> Oct 19 14:01:08 rogue kernel: md: bind<sdc3>
> Oct 19 14:01:08 rogue kernel: md: bind<sda3>
> Oct 19 14:01:08 rogue kernel: raid1: raid set md0 active with 3 out of 4 
> mirrors
> Oct 19 14:01:08 rogue kernel: md: md4 stopped.
> Oct 19 14:01:08 rogue kernel: md: bind<md3>
> Oct 19 14:01:08 rogue kernel: md: bind<md2>
> Oct 19 14:01:08 rogue kernel: md4: setting max_sectors to 128, segment 
> boundary to 32767
> Oct 19 14:01:08 rogue kernel: raid0: looking at md2
> Oct 19 14:01:08 rogue kernel: raid0:   comparing md2(302535936) with 
> md2(302535936)
> Oct 19 14:01:08 rogue kernel: raid0:   END
> Oct 19 14:01:08 rogue kernel: raid0:   ==> UNIQUE
> Oct 19 14:01:08 rogue kernel: raid0: 1 zones
> Oct 19 14:01:08 rogue kernel: raid0: looking at md3
> Oct 19 14:01:08 rogue kernel: raid0:   comparing md3(302535936) with 
> md2(302535936)
> Oct 19 14:01:08 rogue kernel: raid0:   EQUAL
> Oct 19 14:01:08 rogue kernel: raid0: FINAL 1 zones
> Oct 19 14:01:08 rogue kernel: raid0: done.
> Oct 19 14:01:08 rogue kernel: raid0 : md_size is 605071872 blocks.
> Oct 19 14:01:08 rogue kernel: raid0 : conf->hash_spacing is 605071872 
> blocks.
> Oct 19 14:01:08 rogue kernel: raid0 : nb_zone is 1.
> Oct 19 14:01:08 rogue kernel: raid0 : Allocating 8 bytes for hash.
> 
> I add another drive to the failed raid10 device and i get:
> 
> Oct 19 19:25:25 rogue kernel: sd 3:0:0:0: [sdd] 625142448 512-byte 
> hardware sectors (320073 MB)
> Oct 19 19:25:25 rogue kernel: sd 3:0:0:0: [sdd] Write Protect is off
> Oct 19 19:25:25 rogue kernel: sd 3:0:0:0: [sdd] Mode Sense: 00 3a 00 00
> Oct 19 19:25:25 rogue kernel: sd 3:0:0:0: [sdd] Write cache: enabled, 
> read cache: enabled, doesn't support DPO or FU
> A
> Oct 19 19:25:25 rogue kernel: sdd: sdd1
> Oct 19 19:25:28 rogue kernel: sd 3:0:0:0: [sdd] 625142448 512-byte 
> hardware sectors (320073 MB)
> Oct 19 19:25:28 rogue kernel: sd 3:0:0:0: [sdd] Write Protect is off
> Oct 19 19:25:28 rogue kernel: sd 3:0:0:0: [sdd] Mode Sense: 00 3a 00 00
> Oct 19 19:25:28 rogue kernel: sd 3:0:0:0: [sdd] Write cache: enabled, 
> read cache: enabled, doesn't support DPO or FU
> A
> Oct 19 19:25:28 rogue kernel: sdd: sdd1 sdd2 sdd3
> Oct 19 19:26:13 rogue kernel: md: bind<sdd3>
> Oct 19 19:26:13 rogue kernel: RAID1 conf printout:
> Oct 19 19:26:13 rogue kernel: --- wd:3 rd:4
> Oct 19 19:26:13 rogue kernel: disk 0, wo:0, o:1, dev:sda3
> Oct 19 19:26:13 rogue kernel: disk 1, wo:0, o:1, dev:sdb3
> Oct 19 19:26:13 rogue kernel: disk 2, wo:0, o:1, dev:sdc3
> Oct 19 19:26:13 rogue kernel: disk 3, wo:1, o:1, dev:sdd3
> Oct 19 19:26:13 rogue kernel: md: recovery of RAID array md0
> Oct 19 19:26:13 rogue kernel: md: minimum _guaranteed_  speed: 1000 
> KB/sec/disk.
> Oct 19 19:26:13 rogue kernel: md: using maximum available idle IO 
> bandwidth (but not more than 200000 KB/sec) for re
> covery.
> Oct 19 19:26:13 rogue kernel: md: using 128k window, over a total of 
> 9791552 blocks.
> Oct 19 19:26:34 rogue kernel: md: bind<sdd1>
> Oct 19 19:26:34 rogue kernel: RAID1 conf printout:
> Oct 19 19:26:34 rogue kernel: --- wd:3 rd:4
> Oct 19 19:26:34 rogue kernel: disk 0, wo:0, o:1, dev:sda1
> Oct 19 19:26:34 rogue kernel: disk 1, wo:0, o:1, dev:sdb1
> Oct 19 19:26:34 rogue kernel: disk 2, wo:0, o:1, dev:sdc1
> Oct 19 19:26:34 rogue kernel: disk 3, wo:1, o:1, dev:sdd1
> Oct 19 19:26:34 rogue kernel: md: delaying recovery of md1 until md0 has 
> finished (they share one or more physical u
> nits)
> Oct 19 19:27:10 rogue kernel: md: bind<sdd2>
> Oct 19 19:27:10 rogue kernel: RAID1 conf printout:
> Oct 19 19:27:10 rogue kernel: --- wd:1 rd:2
> Oct 19 19:27:10 rogue kernel: disk 0, wo:0, o:1, dev:sdb2
> Oct 19 19:27:10 rogue kernel: disk 1, wo:1, o:1, dev:sdd2
> Oct 19 19:27:10 rogue kernel: md: delaying recovery of md3 until md0 has 
> finished (they share one or more physical u
> nits)
> Oct 19 19:27:22 rogue kernel: ata4: D2H reg with I during NCQ, this 
> message won't be printed again
> Oct 19 19:48:40 rogue kernel: md: md0: recovery done.
> Oct 19 19:48:40 rogue kernel: md: delaying recovery of md3 until md1 has 
> finished (they share one or more physical u
> nits)
> Oct 19 19:48:40 rogue kernel: md: delaying recovery of md1 until md3 has 
> finished (they share one or more physical u
> nits)
> Oct 19 19:48:40 rogue kernel: md: recovery of RAID array md3
> Oct 19 19:48:40 rogue kernel: md: minimum _guaranteed_  speed: 1000 
> KB/sec/disk.
> Oct 19 19:48:40 rogue kernel: md: using maximum available idle IO 
> bandwidth (but not more than 200000 KB/sec) for re
> covery.
> Oct 19 19:48:40 rogue kernel: md: using 128k window, over a total of 
> 302536000 blocks.
> Oct 19 19:48:41 rogue kernel: RAID1 conf printout:
> Oct 19 19:48:41 rogue kernel: --- wd:4 rd:4
> Oct 19 19:48:41 rogue kernel: disk 0, wo:0, o:1, dev:sda3
> Oct 19 19:48:41 rogue kernel: disk 1, wo:0, o:1, dev:sdb3
> Oct 19 19:48:41 rogue kernel: disk 2, wo:0, o:1, dev:sdc3
> Oct 19 19:48:41 rogue kernel: disk 3, wo:0, o:1, dev:sdd3
> Oct 20 09:16:10 rogue kernel: md: md3: recovery done.
> Oct 20 09:16:10 rogue kernel: md: recovery of RAID array md1
> Oct 20 09:16:10 rogue kernel: md: minimum _guaranteed_  speed: 1000 
> KB/sec/disk.
> Oct 20 09:16:10 rogue kernel: md: using maximum available idle IO 
> bandwidth (but not more than 200000 KB/sec) for re
> covery.
> Oct 20 09:16:10 rogue kernel: md: using 128k window, over a total of 
> 240832 blocks.
> Oct 20 09:16:11 rogue kernel: RAID1 conf printout:
> Oct 20 09:16:11 rogue kernel: --- wd:2 rd:2
> Oct 20 09:16:11 rogue kernel: disk 0, wo:0, o:1, dev:sdb2
> Oct 20 09:16:11 rogue kernel: disk 1, wo:0, o:1, dev:sdd2
> Oct 20 09:16:30 rogue kernel: md: md1: recovery done.
> Oct 20 09:16:30 rogue kernel: RAID1 conf printout:
> Oct 20 09:16:30 rogue kernel: --- wd:4 rd:4
> Oct 20 09:16:30 rogue kernel: disk 0, wo:0, o:1, dev:sda1
> Oct 20 09:16:30 rogue kernel: disk 1, wo:0, o:1, dev:sdb1
> Oct 20 09:16:30 rogue kernel: disk 2, wo:0, o:1, dev:sdc1
> Oct 20 09:16:30 rogue kernel: disk 3, wo:0, o:1, dev:sdd1
> 
> The RAID 1 builds perfectly, but five days after that, the system shows a:
> 
> end_request: I/O error, dev sdd, sector 144006110
> raid1: Disk failure on sdd2, disabling device.
> Operation continuing on 1 devices
> end_request: I/O error, dev sdd, sector 144006222
> end_request: I/O error, dev sdd, sector 144268814
> RAID1 conf printout:
> --- wd:1 rd:2
> disk 0, wo:0, o:1, dev:sdb2
> disk 1, wo:1, o:0, dev:sdd2
> RAID1 conf printout:
> --- wd:1 rd:2
> disk 0, wo:0, o:1, dev:sdb2
> 
> a week before i get (under 2.6.18) the following message:
> 
> Oct 10 12:10:56 rogue kernel: ata4.00: exception Emask 0x0 SAct 0x0 SErr 
> 0x0 action 0x2 frozen
> Oct 10 12:10:56 rogue kernel: ata4.00: tag 0 cmd 0xea Emask 0x4 stat 
> 0x40 err 0x0 (timeout)
> Oct 10 12:10:56 rogue kernel: ata4: soft resetting port
> Oct 10 12:10:57 rogue kernel: ata4: softreset failed (1st FIS failed)
> Oct 10 12:10:57 rogue kernel: ata4: softreset failed, retrying in 5 secs
> Oct 10 12:11:02 rogue kernel: ata4: hard resetting port
> Oct 10 12:11:09 rogue kernel: ata4: port is slow to respond, please be 
> patient
> Oct 10 12:11:32 rogue kernel: ata4: port failed to respond (30 secs)
> Oct 10 12:11:32 rogue kernel: ata4: COMRESET failed (device not ready)
> Oct 10 12:11:32 rogue kernel: ata4: hardreset failed, retrying in 5 secs
> Oct 10 12:11:37 rogue kernel: ata4: hard resetting port
> Oct 10 12:11:37 rogue kernel: ata4: SATA link up 3.0 Gbps (SStatus 123 
> SControl 300)
> Oct 10 12:11:37 rogue kernel: ata4.00: configured for UDMA/133
> Oct 10 12:11:37 rogue kernel: ata4: EH complete
> Oct 10 12:11:37 rogue kernel: SCSI device sdd: 625142448 512-byte hdwr 
> sectors (320073 MB)
> Oct 10 12:11:37 rogue kernel: sdd: Write Protect is off
> Oct 10 12:11:37 rogue kernel: sdd: Mode Sense: 00 3a 00 00
> Oct 10 12:11:37 rogue kernel: SCSI device sdd: drive cache: write back
> Oct 10 12:12:54 rogue kernel: ata4.00: exception Emask 0x0 SAct 0x50 
> SErr 0x0 action 0x2 frozen
> Oct 10 12:12:54 rogue kernel: ata4.00: tag 4 cmd 0x61 Emask 0x4 stat 
> 0x40 err 0x0 (timeout)
> Oct 10 12:12:54 rogue kernel: ata4.00: tag 6 cmd 0x61 Emask 0x4 stat 
> 0x40 err 0x0 (timeout)
> Oct 10 12:12:55 rogue kernel: ata4: soft resetting port
> Oct 10 12:13:02 rogue kernel: ata4: port is slow to respond, please be 
> patient
> Oct 10 12:13:25 rogue kernel: ata4: port failed to respond (30 secs)
> Oct 10 12:13:25 rogue kernel: ata4: softreset failed (device not ready)
> Oct 10 12:13:25 rogue kernel: ata4: softreset failed, retrying in 5 secs
> Oct 10 12:13:30 rogue kernel: ata4: hard resetting port
> Oct 10 12:13:37 rogue kernel: ata4: port is slow to respond, please be 
> patient
> Oct 10 12:14:00 rogue kernel: ata4: port failed to respond (30 secs)
> Oct 10 12:14:00 rogue kernel: ata4: COMRESET failed (device not ready)
> Oct 10 12:14:00 rogue kernel: ata4: hardreset failed, retrying in 5 secs
> Oct 10 12:14:05 rogue kernel: ata4: hard resetting port
> Oct 10 12:14:06 rogue kernel: ata4: SATA link up 3.0 Gbps (SStatus 123 
> SControl 300)
> Oct 10 12:14:06 rogue kernel: ata4.00: configured for UDMA/133
> Oct 10 12:14:06 rogue kernel: sd 3:0:0:0: SCSI error: return code = 
> 0x06000000
> Oct 10 12:14:06 rogue kernel: end_request: I/O error, dev sdd, sector 
> 1268542
> Oct 10 12:14:06 rogue kernel: raid1: Disk failure on sdd2, disabling device.
> Oct 10 12:14:06 rogue kernel:   Operation continuing on 1 devices
> Oct 10 12:14:06 rogue kernel: sd 3:0:0:0: SCSI error: return code = 
> 0x06000000
> Oct 10 12:14:06 rogue kernel: end_request: I/O error, dev sdd, sector 
> 1268558
> Oct 10 12:14:06 rogue kernel: ata4: EH complete
> Oct 10 12:14:06 rogue kernel: SCSI device sdd: 625142448 512-byte hdwr 
> sectors (320073 MB)
> Oct 10 12:14:06 rogue kernel: sdd: Write Protect is off
> Oct 10 12:14:06 rogue kernel: sdd: Mode Sense: 00 3a 00 00
> Oct 10 12:14:06 rogue kernel: RAID1 conf printout:
> Oct 10 12:14:06 rogue kernel: SCSI device sdd: drive cache: write back
> Oct 10 12:14:06 rogue kernel: --- wd:1 rd:2
> Oct 10 12:14:06 rogue kernel: disk 0, wo:0, o:1, dev:sdb2
> Oct 10 12:14:06 rogue kernel: disk 1, wo:1, o:0, dev:sdd2
> Oct 10 12:14:06 rogue kernel: RAID1 conf printout:
> Oct 10 12:14:06 rogue kernel: --- wd:1 rd:2
> Oct 10 12:14:06 rogue kernel: disk 0, wo:0, o:1, dev:sdb2
> 
> I have updated from 2.6.18 to 2.6.22 expecting to not have the problem, 
> but the problem remains and i didn't know what could be the problem, the 
> problem  always happen on /dev/sdd, i use LVM on top of the RAID 10 
> software device.
> 
> I am not sure if the problem was because i create the RAID10 using two 
> RAID1 devices and then do a RAID0, or should i have to be used mdadm and 
> the level 10 option ?

It seems to me that if the kernel says there is an io error explicitly
on sdd, then I don't think it really should be any fault with the raid
setup.  My assumption would be that there is a flacky drive, a flacky
cable, or maybe a flacky port on the controller.

It would be interesting to know what happened if you swapped the cables
between sdd and another drive to see if the problem follows the port or
the drive or the cable (you could move sdd's cable to sda, and then
connect sdd to sdc's port and sdc to sdd's port, and if sda then fails,
it should be the cable, if sdd (was sdc) fails, then it must be the
port, and if sdc (was sdd) fails then it must be the drive.  At least
that is what I think.  the raid should deal with moving the drives
around just fine since it should be using the UUIDs for assembly anyhow.

--
Len Sorensen
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists