From owner-freebsd-current Mon Sep 23 18:40: 9 2002 Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id D36E837B401; Mon, 23 Sep 2002 18:40:07 -0700 (PDT) Received: from rwcrmhc53.attbi.com (rwcrmhc53.attbi.com [204.127.198.39]) by mx1.FreeBSD.org (Postfix) with ESMTP id 8174543E4A; Mon, 23 Sep 2002 18:40:07 -0700 (PDT) (envelope-from julian@elischer.org) Received: from InterJet.elischer.org ([12.232.206.8]) by rwcrmhc53.attbi.com (InterMail vM.4.01.03.27 201-229-121-127-20010626) with ESMTP id <20020924014007.QCLI8126.rwcrmhc53.attbi.com@InterJet.elischer.org>; Tue, 24 Sep 2002 01:40:07 +0000 Received: from localhost (localhost.elischer.org [127.0.0.1]) by InterJet.elischer.org (8.9.1a/8.9.1) with ESMTP id SAA37414; Mon, 23 Sep 2002 18:28:28 -0700 (PDT) Date: Mon, 23 Sep 2002 18:28:27 -0700 (PDT) From: Julian Elischer To: Mike Barcroft Cc: walt , freebsd-current@freebsd.org Subject: Re: Question for committers. In-Reply-To: <20020923211028.F14878@espresso.q9media.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=X-UNKNOWN Content-Transfer-Encoding: QUOTED-PRINTABLE Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG On Mon, 23 Sep 2002, Mike Barcroft wrote: > walt writes: > > May I ask a na=EFve question, please? World has been broken at libncur= ses for > > three days. There have been dozens of commits to the -CURRENT tree dur= ing that > > same three days. > >=20 > > How do you committers to -CURRENT keep working when userland is broken?= How > > can you judge the impact of all your changes when you are not rebuildin= g > > the system every day? >=20 > I'm sure most of us aren't installing world each day, only building > and installing certain things, or only building everything with an > older world. >=20 I have 2 machines.. one I commit from and one that is purely updated from a mirror of the cvs tree. After I've build everything relevent and often a whole world, on my editing machine, I commit but it's still possible to make mistakes, For example leving out part of an edit set. Then I wait for it to cycle back to me though the mirror system and then I do a make -k buildworld on the other machine. Hopefully=20 this machine represents "joe developer", and doesn't have teh missing stuff already on it. =20 I go through the log and check for failures.=20 If I consider that that failure isn't anything to do with me, I=20 leave it for whoever made the mistake. If it's my fault (e.g. this morning I failed to commit a pair of edits) I'll catch it and fix it.. cycle time is usually a about 2 hours but sometimes a bit more. (e.g. if you drift off to sleep before it gets back to you :-) The key is the -k together with logging the output. Sometimes I diff consequtive outputs. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message