Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 6 Mar 2001 12:59:55 -0500
From:      lmayhew@freespace.net
To:        freebsd-questions@freebsd.org
Subject:   Hard drive wierdness
Message-ID:  <200103061759.MAA02918@eagle.freespace.net>

next in thread | raw e-mail | index | archive | help
I'm hoping someone can give us some help on an unusual problem.  

Our main server has been running freebsd 3.4 stable for about a year without 
incident.  The problem may have started a few days before Christmas, in that we 
had a problem with our server which resulted in it crashing every time I tried 
to read my mail.  We ended up deleting my inbox, and this appeared to solve the 
problem.  We assumed that I had something corrupt in my mailbox that was 
causing the problem.

A week later, we started getting seekdone errors, which seemed to indicate a 
hard drive failure.  We swapped over to a backup drive, were able to retrieve 
the files from the bad drive with a little work, and replaced the backup drive 
with a new drive.

About a week later, we once again got the seekdone errors, so we got a 
completely new set of hard drives.  We did a new build of freebsd3.4 on each 
drive, then copied over mail, websites, and other relevant files.  Five days 
later, once again we got seekdone errors which seemed to indicate a hard drive 
failure.

Thinking that maybe we had some hardware problem somewhere on the server, we 
built a whole new machine from scratch.  Completely new machine, and the old 
machine was taken off line until we could figure out what was causing the 
problem.

So what happened yesterday?  You guessed it -- hard drive failure in the middle 
of a mail backup.  After running for 3 weeks straight without a problem on the 
new machine, we had yet another hard drive failure, with the same seekdone 
errors showing up.  I swapped the backup drive in and booted from that, but 
when we try to retrieve anything off the original drive, we get a "magic 
Number" error, and we are told that every superblock on the drive is bad.  We 
can't mount the drive to retrieve the last of the info from it.

Incidently, our secondary DNS server has also been running the same version of 
Freebsd without issue for the last year.  (Watch it develop hard drive failures 
now just because I have said that)  The other servers in the office running NT 
and one running an older version of freebsd also don't seem to be affected.  We 
ran a utility on one of the supposed "failed" drives, and it shows no faults in 
the drive.


So, if you're still reading at this point, my questions are:

Does anyone have a clue what could be causing this?  My sanity won't take much 
more...

Anyone have any idea how we can retrieve the last of the information off this 
current drive?  With every superblock supposedly bad we haven't been able to 
mount the file system to retrieve the info.

Any thoughts or suggestions would be most appreciated.

Thanks,

Linda




To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200103061759.MAA02918>