From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dan Carpenter Date: Fri, 07 Jun 2013 08:33:14 +0000 Subject: Re: [patch] ipvs: info leak in __ip_vs_get_dest_entries() Message-Id: <20130607083314.GQ23987@mwanda> List-Id: References: <20130603090049.GA16171@debian> <20130604010327.GA23054@verge.net.au> In-Reply-To: <20130604010327.GA23054@verge.net.au> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Simon Horman Cc: Wensong Zhang , Julian Anastasov , Pablo Neira Ayuso , Patrick McHardy , Jozsef Kadlecsik , "David S. Miller" , netdev@vger.kernel.org, lvs-devel@vger.kernel.org, netfilter-devel@vger.kernel.org, netfilter@vger.kernel.org, coreteam@netfilter.org, kernel-janitors@vger.kernel.org On Mon, Jun 03, 2013 at 06:03:27PM -0700, Simon Horman wrote: > On Mon, Jun 03, 2013 at 12:00:49PM +0300, Dan Carpenter wrote: > > The entry struct has a 2 byte hole after ->port and another 4 byte > > hole after ->stats.outpkts. You must have CAP_NET_ADMIN in your > > namespace to hit this information leak. > > Hi Dan, > > can I verify that it is actually possible to hit this and > thus the patch is a -stable candidate? This is a static checker fix. To me it seems like it's obviously a real info leak. I'm not certain of the impact though. CLONE_NEWNET requires CAP_SYS_ADMIN but on the other hand people are making virtualization products where they give everyone their own namespace with admin privileges. regards, dan carpenter