Mailing-List: contact cygwin-help AT sourceware DOT cygnus DOT com; run by ezmlm List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-owner AT sources DOT redhat DOT com Delivered-To: mailing list cygwin AT sources DOT redhat DOT com To: "Karl M" Cc: cygwin AT cygwin DOT com Subject: Re: OpenSSH broken with cygwin-1.3.2, w32api-20010520 In-Reply-To: Message from "Karl M" of "Tue, 05 Jun 2001 11:23:54 PDT." References: Date: Tue, 05 Jun 2001 18:03:38 -0400 From: Stephen Bailey Message-Id: <20010605220550.C30CA94006@sandmail.sandburst.com> Karl, Thanks for the response. Your patch seems to fix the runaway CPU at the end of the command, and, editorially, I think I've seen similar buggy behavior with select in code that I've ported, but it does not fix the real, initial problem. The initial problem is reported as can't set tty pgrp: not owner. As I mentioned, this problem was introduced between 1.3.1 and 1.3.2. I'd be perfectly happyto take an out-of-the-box solution. I'm using the command to invoke X clients on a remote machine, and I don't want to tie up my shell. Clearly -X is required, and in order to get my shell prompt back, I use -f. I actually would prefer being able to run the ssh command without a Windows console window (or shell) at all, but I have not been able to do that. Any suggestions? For example, I run: % ssh -f -X myhost xterm -T myhost -n myhost & to get an xterm on myhost. With this approach (which worked until 1.3.2) it's still annoying that I can't exit my shell until the xterm has exited. Thanks, Steph -- Want to unsubscribe from this list? Check out: http://cygwin.com/ml/#unsubscribe-simple