Andrew Morton skrev: > On Wed, 10 Oct 2007 10:28:45 +0200 (CEST) > lists@xxxxxxxxx wrote: > >> I get this on brand new hardware, 2xHitachi Deathstar 320gb SATA2 >> (sata_via driver) Sep 28 04:32:40 locker ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen Sep 28 04:32:40 locker ata1.00: cmd b0/d2:f1:00:4f:c2/00:00:00:00:00/00 tag 0 cdb 0x0 data 123392 in Sep 28 04:32:40 locker res 50/00:f1:00:4f:c2/00:00:00:00:00/00 Emask 0x202 (HSM violation) Sep 28 04:32:41 locker current size: 625140335 sectors Sep 28 04:32:41 locker native size: 625142448 sectors Sep 28 04:32:41 locker current size: 625140335 sectors Sep 28 04:32:41 locker native size: 625142448 sectors Another machine: Sep 28 03:47:55 dragonslair ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen Sep 28 03:47:55 dragonslair ata1.00: cmd b0/db:f8:00:4f:c2/00:00:00:00:00/00 tag 0 cdb 0x0 data 126976 in Sep 28 03:47:55 dragonslair res 50/00:f8:00:4f:c2/00:00:00:00:00/00 Emask 0x202 (HSM violation) Sep 28 03:47:55 dragonslair ata1: soft resetting port Sep 28 03:47:55 dragonslair ata1.00: configured for UDMA/133 Sep 28 03:47:55 dragonslair ata1: EH complete Sep 28 03:47:55 dragonslair sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA Sep 28 03:47:55 dragonslair sd 0:0:0:0: [sda] 156250000 512-byte hardware sectors (80000 MB) Sep 28 03:47:55 dragonslair sd 0:0:0:0: [sda] Write Protect is off Sep 28 03:47:55 dragonslair sd 0:0:0:0: [sda] Mode Sense: 00 3a 00 00 Sep 28 03:47:55 dragonslair sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA And yet another: Sep 28 04:33:52 liferaft kernel: ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen Sep 28 04:33:55 liferaft kernel: ata1.00: cmd b0/d2:f1:00:4f:c2/00:00:00:00:00/00 tag 0 cdb 0x0 data 123392 in Sep 28 04:33:55 liferaft kernel: res 50/00:f1:00:4f:c2/00:00:00:00:00/00 Emask 0x202 (HSM violation) Sep 28 04:33:55 liferaft kernel: ata1: soft resetting port Sep 28 04:33:55 liferaft kernel: ata1: SATA link up 3.0 Gbps (SStatus 123 SControl 300) Sep 28 04:33:55 liferaft kernel: ata1.00: configured for UDMA/133 Sep 28 04:33:55 liferaft kernel: ata1: EH complete Another few cases, taken from semi random locations from my log to get the many different data, maybe some of it can help out. Weirdness 1: I have 3 machines, that decide to spew this garbage within the same second? (smartd running at it is around the hour that smartd would run, but it's just this one day Sep 28 that horrible bad) Note 1: Bad/failing hardware creates these type of errors. Note 2: The hardware didn't freeze for me and I believe the freeze is do to swap breaking due to the errors. Note 3: dragonslair's harddisk actually crashed, kernel didn't die, it just remounted read only. Reboot and the disk was missing, more reboot and the machine started with all disks running again, been stable since the 28th Sep. (knock on wood) Note 4: I've changed hardware and kernel in a non controled manner, so I was waiting for another case of these errors where I would be able to write down kernel config. I'm not sure, but I do believe that a keyword with this stuff is SMP and 2.6.22, older kernels doesn't seem to trigger this and non SMP seems to avoid it with 2.6.22, but I can't trigger the error, so there is no way of knowing if the conclusion can be trusted. Dragonslair: P4 x2 3.0 Ghz Chips Intel 865GV & ICH5 32bit SMP kernel (2.6.22) 2 SATA disks WDC WD800JD-75MSA3 (I'm guessing this one has a physical bad disk, since it's the only one the disk has physically failed and the only one with a worrying SMART error: 1 Raw_Read_Error_Rate 0x000f 200 200 051 Pre-fail Always - 5) Locker: AMD 64x2 Chips Nvidia 570 32bit SMP kernel (2.6.22) 6 SATA disks 2xWD3200YS-01PGB0 4xWD3200AAKS-00TMA0 Liferaft: AMD 64x2 Chips Nvidia 590 32bit SMP kernel (vserver 2.6.22 based) 1 SATA disk WD2500KS-00MJB0
Attachment:
smime.p7s
Description: S/MIME Cryptographic Signature