From mboxrd@z Thu Jan 1 00:00:00 1970 From: Nivedita Singhvi Subject: Re: peth0 received packet on own address Date: Mon, 01 May 2006 08:26:14 -0700 Message-ID: <44562896.9050608@us.ibm.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Ian Pratt Cc: xen-devel@lists.xensource.com, Kirk Allan List-Id: xen-devel@lists.xenproject.org Ian Pratt wrote: >>We shouldn't be doing all of those things (peth0) on the >>guest OSs - we had gone to some pains to make sure the MAC >>was unique across the system - have we regressed in that due >>to the -xen kernel or some other reason? > > > It's the internal vifX.X interfaces that have the fe:ff:... addresses. > There really is no business for any of these interfaces to be sourcing > packets at all. > > Is there an option to the 'ip' tool when bringing the interface up to > prevent the ipv6 router solicitation? The ip tool mostly configures the settings on the card, and most of these are network protocol functions which merely get initiated once the interface comes up - they are mostly controlled by sysctl kernel parameters. So, for instance, setting /proc/sys/net/ipv6/conf/$interface/autoconf = 0 will turn off most of that. However, the /proc entry for the device is only created when we register the device, and the MLD and routing functions get initiated only when we designate the interface state as UP. Somewhere in between performing those two steps, we should turn off those sysctl vars. Will test that today.. thanks, Nivedita