From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Tue, 1 May 2001 08:41:47 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Tue, 1 May 2001 08:41:37 -0400 Received: from web5202.mail.yahoo.com ([216.115.106.170]:13070 "HELO web5202.mail.yahoo.com") by vger.kernel.org with SMTP id ; Tue, 1 May 2001 08:41:26 -0400 Message-ID: <20010501124124.1532.qmail@web5202.mail.yahoo.com> Date: Tue, 1 May 2001 05:41:24 -0700 (PDT) From: Rob Landley Subject: New rtl8139 driver prevents ssh from exiting. To: linux-kernel@vger.kernel.org MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org The kernel thread the new rtl8139 driver spawns apparently wants to write to stdout, because it counts as an unfinished process that prevents an ssh session from exiting. I have a script that remotely reconfigures subnets in a vpn, which gets run via an ssh in through eth0 and does, among other things: ifconfig eth1 down ifconfig eth1 up Hence kicking off the thread. I have to run "killall eth1" at the end of the script to make ssh exit. (This doesn't seem to have any negative impact on eth1's functioning.) Anybody feel like shedding light on this? Rob __________________________________________________ Do You Yahoo!? Yahoo! Auctions - buy the things you want at great prices http://auctions.yahoo.com/