2.6.29-rc1-git, pretty reproducible: mount -osync, start write
workload, pull the drive.
Pavel
...
scsi 1:0:0:0: Direct-Access ATA Hitachi HTS54168 SB2O PQ: 0 ANSI: 5
sd 1:0:0:0: [sdb] 156301488 512-byte hardware sectors: (80.0 GB/74.5 GiB)
sd 1:0:0:0: [sdb] Write Protect is off
sd 1:0:0:0: [sdb] Mode Sense: 00 3a 00 00
sd 1:0:0:0: [sdb] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
sd 1:0:0:0: [sdb] 156301488 512-byte hardware sectors: (80.0 GB/74.5 GiB)
sd 1:0:0:0: [sdb] Write Protect is off
sd 1:0:0:0: [sdb] Mode Sense: 00 3a 00 00
sd 1:0:0:0: [sdb] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
sdb: sdb1 sdb2 sdb3
sd 1:0:0:0: [sdb] Attached SCSI disk
sd 1:0:0:0: Attached scsi generic sg1 type 0
kjournald starting. Commit interval 5 seconds
EXT3 FS on sdb1, internal journal
EXT3-fs: recovery complete.
EXT3-fs: mounted filesystem with writeback data mode.
kjournald starting. Commit interval 5 seconds
EXT3 FS on sdb1, internal journal
EXT3-fs: mounted filesystem with writeback data mode.
kjournald starting. Commit interval 5 seconds
EXT3 FS on sdb1, internal journal
EXT3-fs: mounted filesystem with writeback data mode.
kjournald starting. Commit interval 5 seconds
EXT3 FS on sdb1, internal journal
EXT3-fs: mounted filesystem with writeback data mode.
ata2.00: configured for UDMA/100
ata2: EH complete
sd 1:0:0:0: [sdb] 156301488 512-byte hardware sectors: (80.0 GB/74.5 GiB)
sd 1:0:0:0: [sdb] Write Protect is off
sd 1:0:0:0: [sdb] Mode Sense: 00 3a 00 00
sd 1:0:0:0: [sdb] Write cache: disabled, read cache: enabled, doesn't support DPO or FUA
ata2: exception Emask 0x10 SAct 0x0 SErr 0x90200 action 0xe frozen
ata2: irq_stat 0x00400000, PHY RDY changed
ata2: SError: { Persist PHYRdyChg 10B8B }
ata2: hard resetting link
ata2: SATA link down (SStatus 0 SControl 300)
ata2: hard resetting link
ata2: SATA link down (SStatus 0 SControl 300)
ata2: hard resetting link
ata2: SATA link down (SStatus 0 SControl 300)
ata2.00: disabled
ata2: EH complete
sd 1:0:0:0: rejecting I/O to offline device
sd 1:0:0:0: [sdb] Unhandled error code
sd 1:0:0:0: [sdb] Result: hostbyte=0x01 driverbyte=0x00
end_request: I/O error, dev sdb, sector 4991
Buffer I/O error on device sdb1, logical block 616
lost page write due to I/O error on sdb1
ata2.00: detaching (SCSI 1:0:0:0)
ahci: autosuspend disabled
sd 1:0:0:0: [sdb] Stopping disk
sd 1:0:0:0: [sdb] START_STOP FAILED
sd 1:0:0:0: [sdb] Result: hostbyte=0x04 driverbyte=0x00
------------[ cut here ]------------
WARNING: at /data/l/linux/fs/buffer.c:1250 mark_buffer_dirty+0x66/0x80()
Hardware name: ANT
Modules linked in:
Pid: 2970, comm: umount Not tainted 2.6.29-rc1 #263
Call Trace:
[<ffffffff8023da92>] warn_slowpath+0xf2/0x130
[<ffffffff80273502>] find_get_pages+0x32/0xd0
[<ffffffff8027ac67>] pagevec_lookup+0x17/0x20
[<ffffffff8027b378>] truncate_inode_pages_range+0x188/0x360
[<ffffffff80297833>] cache_free_debugcheck+0x103/0x230
[<ffffffff802bcd56>] mark_buffer_dirty+0x66/0x80
[<ffffffff80331249>] ext2_sync_super+0x39/0x60
[<ffffffff803320bc>] ext2_put_super+0xdc/0x100
[<ffffffff8029dfbb>] generic_shutdown_super+0x6b/0x110
[<ffffffff8029e089>] kill_block_super+0x29/0x50
[<ffffffff8029e16f>] deactivate_super+0x5f/0x80
[<ffffffff802b2746>] sys_umount+0x76/0x380
[<ffffffff8022bb72>] ia32_sysret+0x0/0xa
---[ end trace 1a98d19d85775109 ]---
scsi 1:0:0:0: rejecting I/O to dead device
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
On Tue 13-01-09 14:42:05, Pavel Machek wrote:
> 2.6.29-rc1-git, pretty reproducible: mount -osync, start write
> workload, pull the drive.
Looking at the warning more in detail, this is actually a known problem.
When IO fails, the buffer_uptodate flag gets cleared. When we then try to
write the buffer again, we call mark_buffer_dirty() and this warning gets
emitted.
Sadly, fixing this correctly does not seem to be trivial...
Honza
> ------------[ cut here ]------------
> WARNING: at /data/l/linux/fs/buffer.c:1250 mark_buffer_dirty+0x66/0x80()
> Hardware name: ANT
> Modules linked in:
> Pid: 2970, comm: umount Not tainted 2.6.29-rc1 #263
> Call Trace:
> [<ffffffff8023da92>] warn_slowpath+0xf2/0x130
> [<ffffffff80273502>] find_get_pages+0x32/0xd0
> [<ffffffff8027ac67>] pagevec_lookup+0x17/0x20
> [<ffffffff8027b378>] truncate_inode_pages_range+0x188/0x360
> [<ffffffff80297833>] cache_free_debugcheck+0x103/0x230
> [<ffffffff802bcd56>] mark_buffer_dirty+0x66/0x80
> [<ffffffff80331249>] ext2_sync_super+0x39/0x60
> [<ffffffff803320bc>] ext2_put_super+0xdc/0x100
> [<ffffffff8029dfbb>] generic_shutdown_super+0x6b/0x110
> [<ffffffff8029e089>] kill_block_super+0x29/0x50
> [<ffffffff8029e16f>] deactivate_super+0x5f/0x80
> [<ffffffff802b2746>] sys_umount+0x76/0x380
> [<ffffffff8022bb72>] ia32_sysret+0x0/0xa
> ---[ end trace 1a98d19d85775109 ]---
> scsi 1:0:0:0: rejecting I/O to dead device
>
> --
> (english) http://www.livejournal.com/~pavelmachek
> (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
--
Jan Kara <[email protected]>
SUSE Labs, CR