All of lore.kernel.org
 help / color / mirror / Atom feed
From: Dave Chinner <david@fromorbit.com>
To: Linda Walsh <xfs@tlinx.org>
Cc: xfs-oss <xfs@oss.sgi.com>
Subject: Re: xfs file system in process of becoming corrupt; though xfs_repair thinks it's fine! ;-/ (was xfs_dump problem...)
Date: Wed, 30 Jun 2010 09:25:32 +1000	[thread overview]
Message-ID: <20100629232532.GA24712@dastard> (raw)
In-Reply-To: <4C2A749E.4060006@tlinx.org>

On Tue, Jun 29, 2010 at 03:33:02PM -0700, Linda Walsh wrote:
> Dave Chinner wrote:
> >Can you get a list of all the attributes and their sizes on the
> >inodes xfsdump is complaining about?
> >
> >Cheers,
> >
> >Dave.
> Wish I could...had a software problem that had me have to reuse the
> drive I just copied those examples from.
> 
> But have another XFS problem that is much more reliably persistent.
> I don't know if they are at all related, but since I have this problem
> that's a bit "stuck", it's easier to "reproduce".
> 
> Filesystem is one of my larger ones:
> 
> Ishtar:/Torrents> 'df' -T --sync  .
> Filesystem    Type   1K-blocks      Used Available Use% Mounted on
> /dev/dm-7      xfs   6442319744 2305631080 4136688664  36% /Torrents
> Ishtar:/Torrents> 'df' -iT --sync  .
> Filesystem    Type    Inodes   IUsed   IFree IUse% Mounted on
> /dev/dm-7      xfs   1288490112   34313 1288455799    1% /Torrents
> 
> I have 3 files that developed 'bugs' in them in 3 separate
> directories.  Oddly, they were they were 3 copies of the same 3
> files.  Very Odd.
> 
> Symptom is from ls:
> ---------------------------------------------------------------------------------------------------
> Ishtar:/Torrents> 'ls' -ni bad*     ls: cannot access bad/30-Omoide
> to Yakusoku (TV saizu|Reinaʼs Ver.).mp3: No such file or directory
> ls: cannot access bad/31-Omoide to Yakusoku (TV saizu|Tomoeʼs Ver.).mp3: No such file or directory
> ls: cannot access bad/32-Omoide to Yakusoku (TV saizu|Nanualʼs Ver.).mp3: No such file or directory
> bad:
> total 0
> 2359101 ?????????? ? ? ? ?                ? 30-Omoide to Yakusoku (TV saizu|Reinaʼs Ver.).mp3
> 2354946 ?????????? ? ? ? ?                ? 31-Omoide to Yakusoku (TV saizu|Tomoeʼs Ver.).mp3
> 2354949 ?????????? ? ? ? ?                ? 32-Omoide to Yakusoku (TV saizu|Nanualʼs Ver.).mp3
> ls: cannot access bad2/30-Omoide to Yakusoku (TV saizu|Reinaʼs Ver.).mp3: No such file or directory
> ls: cannot access bad2/31-Omoide to Yakusoku (TV saizu|Tomoeʼs Ver.).mp3: No such file or directory
> ls: cannot access bad2/32-Omoide to Yakusoku (TV saizu|Nanualʼs Ver.).mp3: No such file or directory

Those file names have a weird character in them - are you sure that
the terminal supports that character set and is not mangling it and
hence not matching what is actually stored on disk?

> The file system, labeled 'Torrents' is layered on a lvm base (I'm not convinced that
> lvm is as reliable as physical partitions at this point, which is why I mention it).
> 
> It's a 'live' file system.  What info do you want me to dump from it?
> 
> I'm dumping the files from it now, so I can try to remake the file system.
> 
> The problem is 'spreading' to "new" files.  Basically, any file that is being written to
> now, seems to be in danger of becoming inaccessible.

Do those "new" files have the same strange characters in them?

> Also, FWIW -- I did unmount the file system and run xfs_repair.  It
> finds no problems -- so why are these files not accessible?

If there are no problems reported by repair, then I suspect that
it's a terminal level problem...

Cheers,

Dave.
-- 
Dave Chinner
david@fromorbit.com

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

  reply	other threads:[~2010-06-29 23:22 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-06-27  1:10 WARNING xfsdump [still] Cannot allocate memory for list of [root|non-root] attributes for nondir ino xxyz Linda A. Walsh
2010-06-28  2:27 ` Dave Chinner
2010-06-29 22:33   ` xfs file system in process of becoming corrupt; though xfs_repair thinks it's fine! ; -/ (was xfs_dump problem...) Linda Walsh
2010-06-29 23:25     ` Dave Chinner [this message]
2010-06-29 23:55       ` xfs file system in process of becoming corrupt; though xfs_repair thinks it's fine! ;-/ " Michael Weissenbacher
2010-06-30  0:42         ` xfs file system in process of becoming corrupt; though xfs_repair thinks it's fine! ; -/ " Linda A. Walsh
2010-06-30  1:16           ` Dave Chinner
2010-06-30  2:45             ` Linda A. Walsh
2010-07-01 23:58               ` Dave Chinner
2010-07-07  3:18                 ` Linda A. Walsh
2010-07-07  5:56                   ` Linda Walsh
2010-07-07  6:36                     ` Dave Chinner
2010-07-07  9:30                       ` Linda A. Walsh
2010-07-07 21:01                         ` Linda Walsh
2010-06-30  0:01       ` Linda A. Walsh
2010-06-30  1:06         ` xfs file system in process of becoming corrupt; though xfs_repair thinks it's fine! ;-/ " Dave Chinner
2010-06-30  1:52           ` xfs file system in process of becoming corrupt; though xfs_repair thinks it's fine! ; -/ " Linda A. Walsh
2010-06-30 21:01             ` Stan Hoeppner
2010-07-07 21:40               ` utf-8' chars from Winxp machine may be problem related (was Re: xfs file system in process of becoming corrupt; though xfs_repair...) Linda A. Walsh
2010-07-07 23:40                 ` Stan Hoeppner
2010-07-08  0:38                   ` Linda A. Walsh
2010-06-30 18:25     ` xfs file system in process of becoming corrupt; though xfs_repair thinks it's fine! ; -/ (was xfs_dump problem...) Michael Monnerie
2010-06-30 23:30       ` rsync and corrupt inodes (was xfs_dump problem) Dave Chinner
2010-07-01  8:25         ` Michael Monnerie
2010-07-02  2:42           ` Dave Chinner
2010-07-02  6:21             ` Michael Monnerie
2010-07-04 22:53               ` Dave Chinner
2010-07-12 11:28                 ` Michael Monnerie
2010-07-07 21:56         ` Linda Walsh

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20100629232532.GA24712@dastard \
    --to=david@fromorbit.com \
    --cc=xfs@oss.sgi.com \
    --cc=xfs@tlinx.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.