ata*: hard resetting link

[Alexander Fieroch]

Binary package hint: linux-image-2.6.27-7-server

System is stalling every few seconds for 30 seconds each time. Syslog repeats following messages:

[ 4842.388413] ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 frozen
[ 4842.388457] ata1.00: cmd 35/00:08:3f:52:54/00:00:57:00:00/e0 tag 0 dma 4096 out
[ 4842.388460] res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
[ 4842.388531] ata1.00: status: { DRDY }
[ 4842.388557] ata1: hard resetting link
[ 4842.890046] ata1: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
[ 4843.070073] ata1.00: configured for UDMA/33
[ 4843.070094] ata1: EH complete
[ 4843.087693] sd 0:0:0:0: [sda] 1465149168 512-byte hardware sectors (750156 MB)
[ 4843.087693] sd 0:0:0:0: [sda] Write Protect is off
[ 4843.087693] sd 0:0:0:0: [sda] Mode Sense: 00 3a 00 00
[ 4843.128187] sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA

The message is repeating for every hard disk. I'm running a server with a RAID 5 containing 8 SATA disks. The system is ubuntu 8.10 32bit server with kernel 2.6.27-7-server.

The RAID-controller is

02:06.0 SCSI storage controller: Marvell Technology Group Ltd. MV88SX5081 8-port SATA I PCI-X Controller (rev 03)
 Flags: bus master, 66MHz, medium devsel, latency 32, IRQ 27
 Memory at f6000000 (64-bit, non-prefetchable) [size=512K]
 Capabilities: [40] Power Management version 2
 Capabilities: [50] Message Signalled Interrupts: Mask- 64bit+ Queue=0/0 Enable-
 Capabilities: [60] PCI-X non-bridge device
 Kernel driver in use: sata_mv
 Kernel modules: sata_mv

There are 8 SATA hard disks (Seagate Barracuda ES, model ST3750641NS) connected to the RAID controller.

Disabling the write cache for all these hard disks (hdparm -W0 /dev/sd*) is a working solution for me.

[Duncan Sands (baldrick-free)]

I'm having a very similar problem:

Linux version 2.6.27-9-server (similar to original poster)
Ubuntu 8.10 (x86-64; same as original poster, except for 64 bit vs 32 bit)
16 SATA hard disks (WDC WD1500HLFS-0 Rev: 04.0; different disks to original poster)
SCSI storage controller: Marvell Technology Group Ltd. MV88SX6081 8-port SATA II PCI-X Controller (similar to original poster)
Doing raid5 (same as original poster).
Disabling the write cache (as suggested by original poster) also works for me.

Sample message from the logs:
ata1.00: exception Emask 0x0 SAct 0x1 SErr 0x0 action 0x6 frozen
ata1.00: cmd 61/08:00:3f:75:77/00:00:11:00:00/40 tag 0 ncq 4096 out
         res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
ata1.00: status: { DRDY }
ata1: hard resetting link
ata1: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
ata1.00: max_sectors limited to 256 for NCQ
ata1.00: max_sectors limited to 256 for NCQ
ata1.00: configured for UDMA/133
ata1: EH complete
sd 0:0:0:0: [sdc] 293046768 512-byte hardware sectors (150040 MB)
sd 0:0:0:0: [sdc] Write Protect is off
sd 0:0:0:0: [sdc] Mode Sense: 00 3a 00 00
sd 0:0:0:0: [sdc] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA