From owner-freebsd-current  Fri Jul  4 11:39:20 1997
Return-Path: <owner-current>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id LAA03770
          for current-outgoing; Fri, 4 Jul 1997 11:39:20 -0700 (PDT)
Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50])
          by hub.freebsd.org (8.8.5/8.8.5) with SMTP id LAA03764
          for <current@FreeBSD.ORG>; Fri, 4 Jul 1997 11:39:18 -0700 (PDT)
Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id LAA10887; Fri, 4 Jul 1997 11:34:34 -0700
From: Terry Lambert <terry@lambert.org>
Message-Id: <199707041834.LAA10887@phaeton.artisoft.com>
Subject: Re: ppp & HUP.
To: brian@awfulhak.org (Brian Somers)
Date: Fri, 4 Jul 1997 11:34:34 -0700 (MST)
Cc: terry@lambert.org, tom@uniserve.com, chuckr@glue.umd.edu,
        ache@nagual.pp.ru, current@FreeBSD.ORG, brian@awfulhak.org,
        joerg_wunsch@uriah.heep.sax.de
In-Reply-To: <199707040730.IAA01082@awfulhak.demon.co.uk> from "Brian Somers" at Jul 4, 97 08:30:23 am
X-Mailer: ELM [version 2.4 PL24]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-current@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

> > >   Exactly what processes actually exit upon receiving a HUP?  Not many.
> > > Apparently only some user processes.  Daemons NEVER exit, instead they
> > > thrash the system.  Ugh.
> > 
> > All of them that don't explicitly trap HUP.
> > 
> > I always though this should have been handled by revoking the tty's
> > allowing that to HUP to the process group.  This would also mean
> > removing the HUP sending from init.
> 
> Don't confuse the HUP that's sent to everything that's "on"
> in /etc/ttys and the HUP that's sent in death() (before TERM
> & KILL) when the system's coming down.

Everything that's "on" in /etc/ttys has a controlling tty;
they will each get a HUP after the revoke.  The problem here
is the HUP propagation to chile processes of the process group
leader in init.

The HUP that's sent in death() should not be sent.  It should ne
implicit in the revocation of all tty's by shutdown().


					Terry Lambert
					terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.