linux-kernel - Re: overlayfs: BUG in ovl

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20101004173343.GA19733@agent57>
Date:	Mon, 4 Oct 2010 18:33:43 +0100
From:	Andy Whitcroft <apw@...onical.com>
To:	Miklos Szeredi <miklos@...redi.hu>
Cc:	linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: overlayfs: BUG in ovl_whiteout

On Mon, Oct 04, 2010 at 11:23:16AM +0200, Miklos Szeredi wrote:
> On Fri, 1 Oct 2010, Andy Whitcroft wrote:
> > I have been playing with overlayfs for a bit now and there seems to
> > be something a little dodgy when unmounting and remounting an overlay.
> > I modified some files in the overlay, immediatly unmounted the overlay and
> > again immediatly remounted it.  On touching the same files in the overlay
> > I triggered the following BUG (dmesg fragment at the bottom of this email).
> > When trying to reproduce this I also triggered a hard hang.
> 
> Thanks for the bug report, Andy.
> 
> Please change that BUG_ON to a WARN_ON.  This will make this a
> harmless error message in the log instead of a hang.

Applied your patch to move these to WARN_ON.  Now I get a WARN_ON after
the remount.  Now that we don't explode I get the following error from
the triggering command.  I assume that this is the rename which has
triggered the issue:

W: Failed to fetch http://192.168.0.60:3142/archive.ubuntu.com/ubuntu/dists/maverick/Release rename failed, File exists (/var/lib/apt/lists/192.168.0.60:3142_archive.ubuntu.com_ubuntu_dists_maverick_Release -> /var/lib/apt/lists/192.168.0.60:3142_archive.ubuntu.com_ubuntu_dists_maverick_Release).

I note that the file does indeed exist as a real file in both layers.  I
have not as yet managed to generate a naive test that also triggers
this.

-apw
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/