From mboxrd@z Thu Jan 1 00:00:00 1970 From: Daniel Lezcano Subject: Re: tty consideration Date: Tue, 03 Feb 2009 18:15:05 +0100 Message-ID: <49887B99.3010101@free.fr> References: <7B7881568CF40E4388B615CD06F87B980989EA@clara.maurer-it.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <7B7881568CF40E4388B615CD06F87B980989EA-cHQ43lMKbBKenyD9vqZGNUEOCMrvLtNR@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 To: Dietmar Maurer , Linux Containers List-Id: containers.vger.kernel.org Dietmar Maurer wrote: > Another tty problem arises from the implementation of sysv-init/upstart. > Let me explain: > > sysv-init uses /dev/console (or ENV[CONSOLE]), and randomly close/reopen > that device. > > upstart also uses /dev/console, and also randomly close/reopen that > device. > > So far no problem, until udevd is started. udevd rewrites /dev with > reasonable defaults. > > This has the side effect that your tty handling does no longer work. > > Ok, you will argue now that it does not make sense to start udev inside > a container. While this is true, it is incredible difficult to avoid > udev startup, because newer distributions have many dependencies on that > package. > Good catch !