Hello
I have 2 identical machines - both with 3 disks (WDC WD3000HLFS) - root filesystem is under raid1, data partitions are in software raid5 (using mdadm)
linux kernel version - 2.6.25-hardened-r8

About 4 months ago both machines died in the same way - problem with disks - both raids were down, data filesystems was unreachable... (the root filesystem survived)

I thought that it was sth linked with power suply or sth similar - so I made some changes do avoid the problem ...

But few days ago it happened again - at the SAME time - BOTH machines had problems with disks! (again root filesystem survived, data partition was corrupted and raid was unreachable)

I'm thinking about changing the kernel version and add some options to it because now I'm thinking it's ata driver problem ...

But maybe someone can find a source of problem from hardware reports?

It's the report from one of disks:
Last 5 errors are exactly from the time of incident (but i'm afraid those 5 errors are less important than the previous ones) ...

smartctl -a /dev/sda

smartctl version 5.38 [x86_64-pc-linux-gnu] Copyright (C) 2002-8 Bruce Allen
Home page is http://smartmontools.sourceforge.net/                         

=== START OF INFORMATION SECTION ===
Device Model:     WDC WD3000HLFS-01G6U0
Serial Number:    WD-WXL808032081     
Firmware Version: 04.04V01            
User Capacity:    300,069,052,416 bytes
Device is:        Not in smartctl database [for details use: -P showall]
ATA Version is:   8                                                    
ATA Standard is:  Exact ATA specification draft version not indicated  
Local Time is:    Fri Jun 26 10:38:42 2009 CEST                        
SMART support is: Available - device has SMART capability.             
SMART support is: Enabled                                              

=== START OF READ SMART DATA SECTION ===
SMART overall-health self-assessment test result: PASSED

General SMART Values:
Offline data collection status:  (0x84) Offline data collection activity
                                        was suspended by an interrupting command from host.
                                        Auto Offline Data Collection: Enabled.            
Self-test execution status:      (   0) The previous self-test routine completed          
                                        without error or no self-test has ever            
                                        been run.                                         
Total time to complete Offline                                                            
data collection:                 (4800) seconds.                                          
Offline data collection                                                                   
capabilities:                    (0x7b) SMART execute Offline immediate.                  
                                        Auto Offline data collection on/off support.      
                                        Suspend Offline collection upon new               
                                        command.                                          
                                        Offline surface scan supported.                   
                                        Self-test supported.                              
                                        Conveyance Self-test supported.                   
                                        Selective Self-test supported.                    
SMART capabilities:            (0x0003) Saves SMART data before entering                  
                                        power-saving mode.                                
                                        Supports SMART auto save timer.                   
Error logging capability:        (0x01) Error logging supported.                          
                                        General Purpose Logging supported.                
Short self-test routine                                                                   
recommended polling time:        (   2) minutes.                                          
Extended self-test routine                                                                
recommended polling time:        (  59) minutes.                                          
Conveyance self-test routine                                                              
recommended polling time:        (   5) minutes.                                          
SCT capabilities:              (0x303f) SCT Status supported.                             
                                        SCT Feature Control supported.                    
                                        SCT Data Table supported.                         

SMART Attributes Data Structure revision number: 16
Vendor Specific SMART Attributes with Thresholds: 
ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
  1 Raw_Read_Error_Rate     0x000f   200   200   051    Pre-fail  Always       -       0       
  3 Spin_Up_Time            0x0003   195   195   021    Pre-fail  Always       -       3216    
  4 Start_Stop_Count        0x0032   100   100   000    Old_age   Always       -       26      
  5 Reallocated_Sector_Ct   0x0033   200   200   140    Pre-fail  Always       -       0       
  7 Seek_Error_Rate         0x000e   200   200   000    Old_age   Always       -       0       
  9 Power_On_Hours          0x0032   097   097   000    Old_age   Always       -       2480    
 10 Spin_Retry_Count        0x0012   100   253   000    Old_age   Always       -       0       
 11 Calibration_Retry_Count 0x0012   100   253   000    Old_age   Always       -       0       
 12 Power_Cycle_Count       0x0032   100   100   000    Old_age   Always       -       23      
192 Power-Off_Retract_Count 0x0032   200   200   000    Old_age   Always       -       17      
193 Load_Cycle_Count        0x0032   200   200   000    Old_age   Always       -       26      
194 Temperature_Celsius     0x0022   119   107   000    Old_age   Always       -       28      
196 Reallocated_Event_Count 0x0032   200   200   000    Old_age   Always       -       0       
197 Current_Pending_Sector  0x0012   200   200   000    Old_age   Always       -       0       
198 Offline_Uncorrectable   0x0010   200   200   000    Old_age   Offline      -       0       
199 UDMA_CRC_Error_Count    0x003e   200   200   000    Old_age   Always       -       0       
200 Multi_Zone_Error_Rate   0x0008   200   200   000    Old_age   Offline      -       0       

SMART Error Log Version: 1
ATA Error Count: 68 (device log contains only the most recent five errors)
        CR = Command Register [HEX]                                      
        FR = Features Register [HEX]                                     
        SC = Sector Count Register [HEX]                                 
        SN = Sector Number Register [HEX]                                
        CL = Cylinder Low Register [HEX]                                 
        CH = Cylinder High Register [HEX]                                
        DH = Device/Head Register [HEX]                                  
        DC = Device Command Register [HEX]                               
        ER = Error register [HEX]                                        
        ST = Status register [HEX]                                       
Powered_Up_Time is measured from power on, and printed as                
DDd+hh:mm:SS.sss where DD=days, hh=hours, mm=minutes,                    
SS=sec, and sss=millisec. It "wraps" after 49.710 days.                  

Error 68 occurred at disk power-on lifetime: 2453 hours (102 days + 5 hours)
  When the command that caused the error occurred, the device was active or idle.

  After command completion occurred, registers were:
  ER ST SC SN CL CH DH                             
  -- -- -- -- -- -- --                             
  40 51 00 49 b8 f0 40                             

  Commands leading to the command that caused the error were:
  CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
  -- -- -- -- -- -- -- --  ----------------  --------------------
  60 40 00 49 b8 f0 1f 08      00:00:26.823  READ FPDMA QUEUED  
  27 00 00 00 00 00 00 08  49d+17:02:43.547  READ NATIVE MAX ADDRESS EXT
  ec 00 00 00 00 00 00 08  49d+17:02:43.540  IDENTIFY DEVICE           
  ef 03 46 00 00 00 00 08  49d+17:02:43.540  SET FEATURES [Set transfer mode]
  27 00 00 00 00 00 00 08  49d+17:02:43.540  READ NATIVE MAX ADDRESS EXT    

Error 67 occurred at disk power-on lifetime: 2453 hours (102 days + 5 hours)
  When the command that caused the error occurred, the device was active or idle.

  After command completion occurred, registers were:
  ER ST SC SN CL CH DH                             
  -- -- -- -- -- -- --                             
  40 51 00 17 57 00 40                             

  Commands leading to the command that caused the error were:
  CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
  -- -- -- -- -- -- -- --  ----------------  --------------------
  60 28 00 17 57 00 00 08  49d+17:02:43.467  READ FPDMA QUEUED  
  27 00 00 00 00 00 00 08  49d+17:02:43.467  READ NATIVE MAX ADDRESS EXT
  ec 00 00 00 00 00 00 08  49d+17:02:43.460  IDENTIFY DEVICE           
  ef 03 46 00 00 00 00 08  49d+17:02:43.460  SET FEATURES [Set transfer mode]
  27 00 00 00 00 00 00 08  49d+17:02:43.460  READ NATIVE MAX ADDRESS EXT    

Error 66 occurred at disk power-on lifetime: 2453 hours (102 days + 5 hours)
  When the command that caused the error occurred, the device was active or idle.

  After command completion occurred, registers were:
  ER ST SC SN CL CH DH                             
  -- -- -- -- -- -- --                             
  40 51 00 17 57 00 40                             

  Commands leading to the command that caused the error were:
  CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
  -- -- -- -- -- -- -- --  ----------------  --------------------
  60 28 00 17 57 00 00 08  49d+17:02:43.428  READ FPDMA QUEUED  
  27 00 00 00 00 00 00 08  49d+17:02:43.428  READ NATIVE MAX ADDRESS EXT
  ec 00 00 00 00 00 00 08  49d+17:02:43.421  IDENTIFY DEVICE
  ef 03 46 00 00 00 00 08  49d+17:02:43.421  SET FEATURES [Set transfer mode]
  27 00 00 00 00 00 00 08  49d+17:02:43.421  READ NATIVE MAX ADDRESS EXT

Error 65 occurred at disk power-on lifetime: 2453 hours (102 days + 5 hours)
  When the command that caused the error occurred, the device was active or idle.

  After command completion occurred, registers were:
  ER ST SC SN CL CH DH
  -- -- -- -- -- -- --
  40 51 00 17 57 00 40

  Commands leading to the command that caused the error were:
  CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
  -- -- -- -- -- -- -- --  ----------------  --------------------
  60 28 00 17 57 00 00 08  49d+17:02:43.388  READ FPDMA QUEUED
  27 00 00 00 00 00 00 08  49d+17:02:43.388  READ NATIVE MAX ADDRESS EXT
  ec 00 00 00 00 00 00 08  49d+17:02:43.381  IDENTIFY DEVICE
  ef 03 46 00 00 00 00 08  49d+17:02:43.381  SET FEATURES [Set transfer mode]
  27 00 00 00 00 00 00 08  49d+17:02:43.381  READ NATIVE MAX ADDRESS EXT

Error 64 occurred at disk power-on lifetime: 2453 hours (102 days + 5 hours)
  When the command that caused the error occurred, the device was active or idle.

  After command completion occurred, registers were:
  ER ST SC SN CL CH DH
  -- -- -- -- -- -- --
  40 51 00 17 57 00 40

  Commands leading to the command that caused the error were:
  CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
  -- -- -- -- -- -- -- --  ----------------  --------------------
  60 28 00 17 57 00 00 08  49d+17:02:43.349  READ FPDMA QUEUED
  27 00 00 00 00 00 00 08  49d+17:02:43.349  READ NATIVE MAX ADDRESS EXT
  ec 00 00 00 00 00 00 08  49d+17:02:43.342  IDENTIFY DEVICE
  ef 03 46 00 00 00 00 08  49d+17:02:43.342  SET FEATURES [Set transfer mode]
  27 00 00 00 00 00 00 08  49d+17:02:43.342  READ NATIVE MAX ADDRESS EXT

SMART Self-test log structure revision number 1
Num  Test_Description    Status                  Remaining  LifeTime(hours)  LBA_of_first_error
# 1  Conveyance offline  Completed without error       00%         0         -

SMART Selective self-test log data structure revision number 1
 SPAN  MIN_LBA  MAX_LBA  CURRENT_TEST_STATUS
    1        0        0  Not_testing
    2        0        0  Not_testing
    3        0        0  Not_testing
    4        0        0  Not_testing
    5        0        0  Not_testing
Selective self-test flags (0x0):
  After scanning selected spans, do NOT read-scan remainder of disk.
If Selective self-test is pending on power-up, resume after 0 minute delay.


in dmesg I noticed something like this:

ata1.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0                                                                                                                                    
ata1.00: irq_stat 0x40000001                                                                                                                                                                 
ata1.00: cmd ea/00:00:00:00:00/00:00:00:00:00/a0 tag 0                                                                                                                                       
         res 51/04:00:34:cf:f3/00:00:00:f3:40/a3 Emask 0x1 (device error)                                                                                                                    
ata1.00: status: { DRDY ERR }                                                                                                                                                                
ata1.00: error: { ABRT }                                                                                                                                                                     
ata1.00: configured for UDMA/133                                                                                                                                                             
ata1: EH complete                                                                                                                                                                            
sd 0:0:0:0: [sda] 586072368 512-byte hardware sectors (300069 MB)                                                                                                                            
sd 0:0:0:0: [sda] Write Protect is off                                                                                                                                                       
sd 0:0:0:0: [sda] Mode Sense: 00 3a 00 00                                                                                                                                                    
sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA                                                                                                      
ata1.00: exception Emask 0x0 SAct 0x1 SErr 0x0 action 0x0                                                                                                                                    
ata1.00: irq_stat 0x40000008                                                                                                                                                                 
ata1.00: cmd 60/08:00:df:0f:85/00:00:00:00:00/40 tag 0 ncq 4096 in                                                                                                                           
         res 41/40:00:df:0f:85/24:00:00:00:00/40 Emask 0x409 (media error) <F>                                                                                                               
ata1.00: status: { DRDY ERR }                                                                                                                                                                
ata1.00: error: { UNC }                                                                                                                                                                      
ata1.00: configured for UDMA/133                                                                                                                                                             
ata1: EH complete                                                                                                                                                                            
sd 0:0:0:0: [sda] 586072368 512-byte hardware sectors (300069 MB)                                                                                                                            
sd 0:0:0:0: [sda] Write Protect is off                                                                                                                                                       
sd 0:0:0:0: [sda] Mode Sense: 00 3a 00 00                                                                                                                                                    
sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA            

Thanks for any help!

regards
nichu