mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Andreas Dilger <adilger@clusterfs.com>
To: Felix von Leitner <felix-kernel@fefe.de>
Cc: linux-kernel@vger.kernel.org
Subject: Re: corrupt ext3 that even debugfs refuses to repair
Date: Thu, 22 May 2003 09:09:12 -0600	[thread overview]
Message-ID: <20030522090912.J4198@schatzie.adilger.int> (raw)
In-Reply-To: <20030522132646.GB1506@codeblau.de>; from felix-kernel@fefe.de on Thu, May 22, 2003 at 03:26:46PM +0200

On May 22, 2003  15:26 +0200, Felix von Leitner wrote:
> I just found out that my CVS ext3 file system is corrupted.
> The symptom is that du complains about "file not found", i.e. getdents64
> returns the files but stat says they aren't there.
> 
> e2fsck finds nothing wrong with the filesystem.
> 
> I isolated the files into one directory which I moved to /broken.  This
> is what debugfs says:
> 
> # debugfs -w /dev/discs/disc0/part5
> debugfs 1.33 (21-Apr-2003)
> debugfs:  cd /broken
> debugfs:  ls
>  6243596  (12) .    2  (4084) ..    0  (4096) nsISO88592ToUnicode.cpp
>  0  (4096) nsKOI8UToUnicode.o    0  (4096) nsUnicodeToMacDevanagari.h
>  0  (4096) nsUnicodeToZapfDingbat.h    0  (620) nsCP1258ToUnicode.h
>  6243907  (3476) .cvsignore    0  (4096) nsMacUkrainianToUnicode.cpp
> debugfs:  rm foo
> rm: File not found by ext2_lookup while trying to resolve filename
> debugfs:  ls
> ls: invalid option -- o
> zsh: 5846 segmentation fault  debugfs -w /dev/discs/disc0/part5
> 
> 
> Huh?  debugfs does not appear to be a very stable tool ;)

Patches are welcome, since you are probably the only person with a
filesystem corrupted in exactly the right way...  At least run "e2image"
on this partition before throwing it away...

> Once again, with nsCP1258ToUnicode.h this time:
> 
> debugfs:  rm nsCP1258ToUnicode.h
> rm: File not found by ext2_lookup while trying to resolve filename
> debugfs:  unlink nsCP1258ToUnicode.h
> unlink_file_by_name: No free space in the directory

These files are already deleted according to the debugfs output above
(ino == 0 means they are deleted).  Only "." and ".cvsignore" have
valid inode numbers.

> What can I do now (besides backup and restore, obviously)

It is possible that these inodes are in lost+found, but that is only
a guess.

Cheers, Andreas
--
Andreas Dilger
http://sourceforge.net/projects/ext2resize/
http://www-mddsp.enel.ucalgary.ca/People/adilger/


  reply	other threads:[~2003-05-22 14:58 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-05-22 13:26 Felix von Leitner
2003-05-22 15:09 ` Andreas Dilger [this message]
2003-05-25 21:51 ` Theodore Ts'o

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=20030522090912.J4198@schatzie.adilger.int \
    --to=adilger@clusterfs.com \
    --cc=felix-kernel@fefe.de \
    --cc=linux-kernel@vger.kernel.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®