public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Calvin Spealman <calvin@ironfroggy.com>
To: linux-kernel@vger.kernel.org
Subject: Re: Filesystem space accounting bug
Date: Wed, 26 May 2004 07:38:24 +0000	[thread overview]
Message-ID: <1983750.MKsGDOebfy@ironfroggy.com> (raw)
In-Reply-To: 20040525150512.GE7195@secretlab.mine.nu

I had a similar problem a few days ago. My wife was working in GIMP and
called me in, complaining about an Out Of Space error. I didn't find it out
of the ordinary, so I removed lots of temp files and tarballs of things
already installed. I cleared up a few gigs.

A few hours later, she got the same errors. I deleted all the videos I
already had backed-up on CD, since my DVD player can play those anyway.

A few hours later, she got the same errors. I knew at this point something
weird was going on. A few friends suggested my system had been compromised,
but I knew better. Eventually I tried rebooting to a Live CD, so I could
run fsck.ext3, but everything now looked fine. I haven't had the problem
since, and I have no idea what caused it.

Walter Hofmann wrote:

> Hi,
> 
> I have a backup script which creates a filesystem image which is later
> written to DVD. It is created & loop-mounted with
> 
>  dd if=/dev/zero of=${DARIMAGE} bs=1024k count=4460 || exit 1
>  mke2fs -F -b 4096 -m 0 -N 32 -O sparse_super -L BACKUP ${DARIMAGE} ||
>  exit 1 sync
>  mount /mnt/dar
>  touch /mnt/dar/backup-stamp
> 
> Then I use the "dar" utility to create a single large file on this. I
> believe, but have not checked, that dar writes the file linearly. dar is
> instructed to stop writing to the filesystem before it reaches 4GB.
> 
> However, recently, dar failed after approx. 3GB with ENOSPC. The
> filesystem was still mounted, so I could check that there really were
> only 3GB written to it. Still, I could not even create a 1 byte file on
> it, although there should be around 1GB free space left.
> 
> There were no error messages logged. I unmounted the image and ran
> e2fsck on it and it reported that the free block count in a number of
> groups on the filesystem was wrong.
> 
> I'm using Linux 2.6.6. I tried to reproduce this, but now it works
> again.
> 
> Walter

-- 
 


      reply	other threads:[~2004-05-27  1:00 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-05-25 15:05 Filesystem space accounting bug Walter Hofmann
2004-05-26  7:38 ` Calvin Spealman [this message]

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=1983750.MKsGDOebfy@ironfroggy.com \
    --to=calvin@ironfroggy.com \
    --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