From mboxrd@z Thu Jan 1 00:00:00 1970 From: Darren Tucker Subject: Re: [Bugme-new] [Bug 1360] New: Can't access /proc/self/fd/0 from sshd when no pty allocated. Date: Wed, 22 Oct 2003 18:40:51 +1000 Sender: netdev-bounce@oss.sgi.com Message-ID: <3F964293.FA2C3A7A@zip.com.au> References: <200310151142.h9FBgi5k029003@fire-1.osdl.org> <20031018172736.417bf89d.akpm@osdl.org> <20031018235735.52f5e15a.davem@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7BIT Cc: Andrew Morton , netdev@oss.sgi.com Return-path: To: "David S. Miller" Errors-to: netdev-bounce@oss.sgi.com List-Id: netdev.vger.kernel.org "David S. Miller" wrote: > > On Sat, 18 Oct 2003 17:27:36 -0700 > Andrew Morton wrote: > > > This appears to be deliberate: [snip] > > > > I do not know why; the comment is fairly useless. > > For many socket types there are no sane open() semantics. > > If I open() a unconnected socket, what does that mean? > Should it connect, of do something like a dup()? > > This code has been this way for a long time, anyone who wants to > change this needs to deal with and address all the aforementioned > issues. Thanks for looking at this. I've done some digging on the OpenSSH side: originally sshd used pipes on Linux but they were found [0] to have problems that cause lockups in (at least) rsync. [0] http://marc.theaimsgroup.com/?l=openssh-unix-dev&m=94914935131492 -- Darren Tucker (dtucker at zip.com.au) GPG key 8FF4FA69 / D9A3 86E9 7EEE AF4B B2D4 37C9 C982 80C7 8FF4 FA69 Good judgement comes with experience. Unfortunately, the experience usually comes from bad judgement.