[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: SCSI Issues? Are all of my disks bad?

On 2016-06-10 20:17, alexmcwhirter@triadic.us wrote:

On 2016-06-10 19:27, alexmcwhirter@triadic.us wrote:

Using the qlogicpti driver i seem to get these issue when writing data
to an ext4 disk. I've tried three different disks. Im on kernel 4.6.1,
but it also happens on 4.5.2

[  700.648449] scsi_io_completion: 18 callbacks suppressed
[  700.714696] sd 0:0:0:0: [sda] tag#0 FAILED Result:
hostbyte=DID_ERROR driverbyte=DRIVER_OK
[  700.817404] sd 0:0:0:0: [sda] tag#0 CDB: Write Same(10) 41 00 07 48
88 dd 00 00 08 00
[  700.914983] blk_update_request: 18 callbacks suppressed
[ 700.981195] blk_update_request: I/O error, dev sda, sector 122194141 
[  701.136856] sd 0:0:0:0: [sda] tag#0 FAILED Result:
hostbyte=DID_ERROR driverbyte=DRIVER_OK
[  701.239781] sd 0:0:0:0: [sda] tag#0 CDB: Write Same(10) 41 00 07 48
49 85 00 00 08 00
[ 701.337254] blk_update_request: I/O error, dev sda, sector 122177925 
[  701.499375] sd 0:0:0:0: [sda] tag#0 FAILED Result:
hostbyte=DID_ERROR driverbyte=DRIVER_OK
[  701.602233] sd 0:0:0:0: [sda] tag#0 CDB: Write Same(10) 41 00 07 4a
78 b5 00 00 08 00
[ 701.699702] blk_update_request: I/O error, dev sda, sector 122321077 
[  701.852861] sd 0:0:0:0: [sda] tag#0 FAILED Result:
hostbyte=DID_ERROR driverbyte=DRIVER_OK
[  701.955608] sd 0:0:0:0: [sda] tag#0 CDB: Write Same(10) 41 00 07 4a
78 bd 00 00 08 00
[ 702.053063] blk_update_request: I/O error, dev sda, sector 122321085 
[  702.206362] sd 0:0:0:0: [sda] tag#0 FAILED Result:
hostbyte=DID_ERROR driverbyte=DRIVER_OK
[  702.309112] sd 0:0:0:0: [sda] tag#0 CDB: Write Same(10) 41 00 07 4a
78 c5 00 00 08 00
[ 702.406616] blk_update_request: I/O error, dev sda, sector 122321093 
[  702.559887] sd 0:0:0:0: [sda] tag#0 FAILED Result:
hostbyte=DID_ERROR driverbyte=DRIVER_OK
[  702.662594] sd 0:0:0:0: [sda] tag#0 CDB: Write Same(10) 41 00 07 4a
78 cd 00 00 08 00
[ 702.760102] blk_update_request: I/O error, dev sda, sector 122321101 
[  702.913392] sd 0:0:0:0: [sda] tag#0 FAILED Result:
hostbyte=DID_ERROR driverbyte=DRIVER_OK
[  703.016148] sd 0:0:0:0: [sda] tag#0 CDB: Write Same(10) 41 00 07 4a
78 d5 00 00 08 00
[ 703.113574] blk_update_request: I/O error, dev sda, sector 122321109 
[  703.269364] sd 0:0:0:0: [sda] tag#0 FAILED Result:
hostbyte=DID_ERROR driverbyte=DRIVER_OK
[  703.372378] sd 0:0:0:0: [sda] tag#0 CDB: Write Same(10) 41 00 07 48
be fd 00 00 08 00
[ 703.469873] blk_update_request: I/O error, dev sda, sector 122207997 
[  703.622830] sd 0:0:0:0: [sda] tag#0 FAILED Result:
hostbyte=DID_ERROR driverbyte=DRIVER_OK
[  703.725530] sd 0:0:0:0: [sda] tag#0 CDB: Write Same(10) 41 00 07 48
bf 05 00 00 08 00
[ 703.823051] blk_update_request: I/O error, dev sda, sector 122208005 
[  703.988346] sd 0:0:0:0: [sda] tag#0 FAILED Result:
hostbyte=DID_ERROR driverbyte=DRIVER_OK
[  704.091153] sd 0:0:0:0: [sda] tag#0 CDB: Write Same(10) 41 00 07 48
bf 0d 00 00 08 00
[ 704.188709] blk_update_request: I/O error, dev sda, sector 122208013 
[  705.762080] scsi_io_completion: 18 callbacks suppressed

It's odd that the bad sectors are all place 8 sectors apart from each
other. Granted i've only tried fujitsu branded disks. I'll spin up
some hitachi disks...



The same issue occurs on my hitachi disk, except this time it only
happens every 4096 sectors. Another weird number...

[ 1146.706829] sd 0:0:1:0: [sdb] tag#0 CDB: Write Same(10) 41 00 00 00
41 50 00 10 00 00
[ 1146.804343] blk_update_request: I/O error, dev sdb, sector 16720
[ 1157.503807] sd 0:0:1:0: [sdb] tag#0 FAILED Result:
hostbyte=DID_ERROR driverbyte=DRIVER_OK
[ 1157.606581] sd 0:0:1:0: [sdb] tag#0 CDB: Write Same(10) 41 00 00 00
51 50 00 10 00 00
[ 1157.704058] blk_update_request: I/O error, dev sdb, sector 20816

It should be noted that this does not occur on Gentoo. Granted, on
gentoo im forced to use kernel 3.18 as gentoo requires rsync for it's
package manager and rsync is broken on sparc 3.19+. It seems the
Debian 4.6.1 kernel doesn't have this rsync issue. So maybe i should
give kernel 4.6 a try on Gentoo and try to nail down whats happening
here with the I/O.


Sorry for all the list spam lately...
Anyways to add to this some more, this issue only occurs with ext4. i tested btrfs and ext2 fairly extensively with no dmesg errors. Not sure if this is mkfs not laying down the FS correctly or the kernel code misbehaving. 

And now im off to build an ISO with btrfs-progs built in.
Reply to: