[Wolves] Disk errors

Ron Wellsted ron at wellsted.org.uk
Mon May 2 17:35:37 BST 2005


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Adam Sweet wrote:
> --- Ron Wellsted <ron at wellsted.org.uk> wrote:
> 
>>-----BEGIN PGP SIGNED MESSAGE-----
>>Hash: SHA1
>>
>>Adam Sweet wrote:
>>
>>>During the process of my last post, I noticed a
>>
>>nasty
>>
>>>set of disk errors in my logs:
>>>
>>>hda: dma_intr: status=0x51 { DriveReady
>>
>>SeekComplete
>>
>>>Error }
>>>hda: dma_intr: error=0x40 { UncorrectableError },
>>>LBAsect=48078696, high=2, low= 14524264,
>>>sector=48078696
>>>
>>>Thankfully this is my Windows partition that I
>>
>>don't
>>
>>>use anymore but I was going to swap my /home
>>
>>partition
>>
>>>to it as it's a 5 month old 160GB disk. Am I about
>>
>>to
>>
>>>lose the disk? Or would a format solve the issue,
>>>meaning I can format it as ext3 (it's NTFS at the
>>
>>mo)
>>
>>>and not worry about losing data?
>>>
>>>Does anyone know if these are filesystem specific
>>>errors, or are they the beginning of a hardware
>>>failure? From my googling it sounds like hardware
>>>errors, I'd just prefer a more experienced
>>
>>opinion.
>>
>>>Thanks
>>
>>Try a "dd if=/dev/hda of=/dev/null bs=16k" and see
>>if the error occurs
>>again.  If it does again at the same location, you
>>have a hard error on
>>the disk at that location.. You can work around it
>>with the mkfs -c
>>option which will map out the bad blocks.
> 
> 
> The example error was just one of many. dd crapped out
> with:
> 
> dd: reading `/dev/hda': Input/output error
> 
> 
>>If you get the same sort of error in a different
>>location, you may have
>>a disk drive that is on the way out.  Take a backup
>>of everything important ASAP.
> 
> 
> Yeah there are a whole list of these, about 70 lines,
> most of them seem to be at the same 4 sectors.
> 
> 
>>smartd might be useful here to monitor the drive. ;)
> 
> 
> I'll back up my stuff, it's not all important. I think
> I'll have a look at the returns policy too as it's
> only about 5 months old.
> 
> Could this have been caused by heat? I lost another
> disk around November/December and this was the
> replacement. I noticed recently that I was having heat
> problems and the disks were just about touchable. the
> other problems have been resolved since I took the
> side off the case.
> 
> What a pain in the arse.
> 
> Thanks Ron.
> 
> Ad
> 

Could be heat.  A couple of years ago we tried out some 15,000 rpm
drives.  Within 6 months we had replaced the whole lot because they ran
hot and kept failing :(

Makes you appreciate the value of RAID 1 or 5 and good backups!


- --
Ron Wellsted
http://www.wellsted.org.uk
ron at wellsted.org.uk
FWD:519961  Gossiptel:9309811
N 52.567623, W 2.137621
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iQEVAwUBQnZWMEtP/KMNOfRbAQJi+wf+KpbH/jd6dSoLBOenMCfzUGAtVjzS7GI1
AVT7IWxQUktkFlO3/T7M8bqrPDkzyK38BTx1vM0tYja+FSfaf1kw9SVWZrVjklMM
kKv5V4kCk8td/qo434jAcYdwvtG+CpP28shtXi6VL7KOF6IMe6VVW4DZJbzFta2F
VTrLkj22bTvG9dFBM/MSkDlNoBGB+VAeEYPZg7RC0viJ1YDaqkGt7Y0GBdnQMlwR
W1UFw6AXw7cYB8d2/+6GhcyseievDOlpJJWVMvzmiH+JQuHGinVVl/QZ9/4fFmoD
YbhLgJHCZxSioqeYUlROK6UZ6ICpb1/RsK/7qx3hKfldfS08yiDSBQ==
=+Sup
-----END PGP SIGNATURE-----



More information about the Wolves mailing list