kernel hdd error msg

[Jeff Vian]

Laurence Orchard wrote:

>Hi
>
>I get these all the time on one of my IBM disks, accompanied by a
>chugging sound as the disk retries something. While it is doing this the
>machine will not do anything else!
>  
>
If this is repeatedly occurring on the same drive it likely indicates a 
failing drive.

>It occurs on particular files in the file system which suggests there is
>some error actually there. To clear the problem I copy the
>whole partition to another disk then remake the file system using:
>mkfs -t ext3 -c -c /dev/hde1
>
>then copy everything back.
>
>this uses a destructive scan for badblocks on the partition.
>just using 'badblocks' without rebuilding the FS does not seem to work
>the same.
>This will solve the problem for about 6 to 9 months.
>
>  
>
This seems to indicate a possible problem with the drive, since the 
destructive test will identify and mark any bad blocks and then not use 
them.  However, if there is a failing area on the disk, it would not be 
seen or grow except when that area on the disk where it is physically 
located is used.  Thus, if you can isolate the drive area where it is 
located you might be able to avoid it, but I would replace the drive.

Drives are cheap nowdays. (I can buy 18gb scsi drives for about $10  today)

>I would be interested to know if this is just a glitch and not a real
>error, as I am considering junking the disk in question.
>
>If it is just a glitch? how do I stop it retrying all the time?
>
>Laurence
>
>
>On Wed, 2004-02-25 at 15:40, Dave Jones wrote:
>  
>
>>On Wed, 2004-02-25 at 15:36, Adam Voigt wrote:
>>    
>>
>>>I've been getting these on a couple of my older boxes I use for testing,
>>>it's quite annoying when your on the console. Any way to silence these
>>>errors?
>>>
>>>
>>>On Wed, 2004-02-25 at 10:16, Dave Jones wrote:
>>>      
>>>
>>>>On Wed, 2004-02-25 at 06:48, david walcroft wrote:
>>>>        
>>>>
>>>>>Is this error saying the disks are dying.
>>>>>
>>>>>Feb 24 16:03:50 reddwarf kernel: hdb: drive_cmd: status=0x51 { 
>>>>>DriveReady SeekComplete Error }
>>>>>
>>>>>Feb 24 16:03:50 reddwarf kernel: hdb: drive_cmd: error=0x04 { 
>>>>>DriveStatusError }
>>>>>
>>>>>Feb 24 16:14:50 reddwarf kernel: hdc: drive_cmd: status=0x51 { 
>>>>>DriveReady SeekComplete Error }
>>>>>
>>>>>Feb 24 16:14:50 reddwarf kernel: hdc: drive_cmd: error=0x04 { 
>>>>>DriveStatusError }
>>>>>          
>>>>>
>>>>harmless.
>>>>51/04 is the drive saying Eh? in response to the kernel
>>>>asking if it supports a certain ata command.  It's not uncommon to
>>>>see these with older drives.
>>>>        
>>>>
>>they are largely useless in this case, and cause more panic than
>>anything else. I'll have a word with the IDE bod.
>>
>>	Dave
>>
>>    
>>
>
>
>  
>