To: rsc@swtch.com
CC: akpm@linux-foundation.org, linux-fsdevel@vger.kernel.org,
       linux-kernel@vger.kernel.org
In-reply-to: <ee9e417a0702091623t1ccf55b8wb05b64b9c51474a6@mail.gmail.com>
	(rsc@swtch.com)
Subject: Re: [PATCH] fix quadratic behavior of shrink_dcache_parent()
References: <E1HFdnm-0006DB-00@dorka.pomaz.szeredi.hu>
	 <20070209160055.bbd20c9f.akpm@linux-foundation.org> <ee9e417a0702091623t1ccf55b8wb05b64b9c51474a6@mail.gmail.com>
Message-Id: <E1HFnsK-0007Ep-00@dorka.pomaz.szeredi.hu>
From: Miklos Szeredi <miklos@szeredi.hu>
Date: Sat, 10 Feb 2007 09:46:28 +0100
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1209
Lines: 30

> > "The file system mounted on /tmp/z in the example contains 2^50
> > directories".   heh.
> >
> > I do wonder how realistic this problem is in real life.
> 
> That's a fair concern, although I was trying this as part
> of evaluating how much someone could hose a system
> if we let them mount arbitrary FUSE servers.  And the
> answer is: they could make it completely unusable,
> requiring reboot.
> 
> I ran a later test that printed how deep it got into
> the file tree and it was only a few hundred thousand
> if I recall correctly.  A determined attacker might even
> manage to do this in a normal file system.

Unfortunately this patch doesn't completely solve this problem, since
the system will still be hosed due to all memory being used up by
dentries.  And I bet the OOM killer won't find the real target (du)
but will kill anything before that.

So the second part of the problem is to somehow limit the number of
dentries used.  Not easy...

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