[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: ext3 device reported to be 100% full, but we do not know where? - solved

On Tue, Jul 23, 2002 at 10:55:41PM +0200, Michael Hoennig wrote:
> ok, then at least lsof should report such files.  Actually I forgot
> something: Even after all processes holding this file handle, the storage
> was never freed, only after umount/mount!

In my experience, lsof has at times been broken on Linux.  This is no
wonder, as lsof supports lots of *nix systems, and some of the interfaces
are fragile.  Also, be sure to run it as root; depending on how things
are installed (e.g., which kernel, suid bits), you may not see process info
that you can't access in /proc.

The simplest way to see what processes are standing on a filesystem is
to use (as root)

  /sbin/fuser -m <mntpoint>

This is from the (Linux-specific) psmisc package.  See fuser(1).
for details.

> I have not checked it yet, but
> if this storage is not even counted in quota, which I guess is the case,
> it is a possible local DoS.  Any user, even with a reasonable quota, could
> flood the device.

The quota is associated with the owner of the inode, which doesn't change
just because the file has no links in the filesystem.  When the inode is
release after the last close, and the storage is returned, the quota will be
credited to the user.


   Bill Rugolsky

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]