From owner-freebsd-hackers Wed Jul 24 18:56:05 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id SAA02080 for hackers-outgoing; Wed, 24 Jul 1996 18:56:05 -0700 (PDT) Received: from panda.hilink.com.au (panda.hilink.com.au [203.2.144.5]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id SAA02044 for ; Wed, 24 Jul 1996 18:55:40 -0700 (PDT) Received: (from danny@localhost) by panda.hilink.com.au (8.7.5/8.7.3) id LAA03576; Thu, 25 Jul 1996 11:54:35 +1000 (EST) Date: Thu, 25 Jul 1996 11:54:34 +1000 (EST) From: "Daniel O'Callaghan" To: Hans Zuidam cc: freebsd-hackers@freebsd.org Subject: Re: catching /etc/rc output In-Reply-To: <199607241503.RAA01095@truk.brandinnovators.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Wed, 24 Jul 1996, Hans Zuidam wrote: > Is there a convenient way to direct all output from the /etc/rc scripts > to both the console and a file other than rewriting them? Scrollback > is not always sufficient especially after lengthy fsck output. How about starting by building a statically linked version of tee(1). Pop it into /bin and wedge it into /etc/rc where appropriate. Trouble is that you don't have anywhere to write the file unless you (a) brute-force fsck of / and write on that, or (b) have a special partition wd0h, 1MB in size, which you mount for the purposes of catching fsck output, then umount when finished. Thus wd0h is almost guaranteed to be clean and writable. I've given myself a project of making it straightforward to get a two-stage boot process so a machine which is administered remotely can come half-way up enough to be telnetted to. Danny