From mboxrd@z Thu Jan 1 00:00:00 1970 From: John Dykstra Subject: Re: [PATCH 2/2] c/r: Add AF_INET support (v3) Date: Wed, 29 Jul 2009 17:10:29 -0500 Message-ID: <1248905429.7129.22.camel@Maple> References: <1246994776-1882-1-git-send-email-danms@us.ibm.com> <1246994776-1882-3-git-send-email-danms@us.ibm.com> <1248555754.16746.55.camel@merlyn> <874oswer21.fsf@caffeine.danplanet.com> <4A6F306A.40303@librato.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Cc: containers-qjLDD68F18O7TbgM5vRIOg@public.gmane.org, netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Dan Smith , Alexey Dobriyan To: Oren Laadan Return-path: In-Reply-To: <4A6F306A.40303-RdfvBDnrOixBDgjK7y7TUQ@public.gmane.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: containers-bounces-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org Errors-To: containers-bounces-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org List-Id: netdev.vger.kernel.org On Tue, 2009-07-28 at 13:07 -0400, Oren Laadan wrote: > When you destroy the virtual interface after (or during) the > > migration, the TIME_WAIT socket will disappear from the sending > host. > > Note that such sockets are unlikely to be referenced by _any_ process > because they will have been closed already. So the checkpoint code > will need to find such sockets not through file descriptors. Dan, to maybe save you some searching... The time-wait chains can be accessed via tcp_hashinfo. -- John