From owner-freebsd-current@FreeBSD.ORG Thu Jun 26 08:55:16 2003 Return-Path: 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 CEE9437B401 for ; Thu, 26 Jun 2003 08:55:16 -0700 (PDT) Received: from postal3.es.net (postal3.es.net [198.128.3.207]) by mx1.FreeBSD.org (Postfix) with ESMTP id 1BFE643F75 for ; Thu, 26 Jun 2003 08:55:16 -0700 (PDT) (envelope-from oberman@es.net) Received: from ptavv.es.net ([198.128.4.29]) by postal3.es.net (Postal Node 3) with ESMTP (SSL) id MUA74016; Thu, 26 Jun 2003 08:55:13 -0700 Received: from ptavv (localhost [127.0.0.1]) by ptavv.es.net (Tachyon Server) with ESMTP id 5112C5D08; Thu, 26 Jun 2003 08:55:13 -0700 (PDT) To: Tobias Roth In-Reply-To: Message from Tobias Roth <20030626101942.GA15745@speedy.unibe.ch> Date: Thu, 26 Jun 2003 08:55:13 -0700 From: "Kevin Oberman" Message-Id: <20030626155513.5112C5D08@ptavv.es.net> cc: Christoph Kukulies cc: current@freebsd.org Subject: Re: world build fails since yesterday X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 26 Jun 2003 15:55:17 -0000 > Date: Thu, 26 Jun 2003 12:19:42 +0200 > From: Tobias Roth > Sender: owner-freebsd-current@freebsd.org > > Hi > > I get the same behaviour on my T30. Although the indications for a > hardware problem are very strong, I am not yet convinced that this > really is one. > > My suspicion is that there are problems with current (as well with > 5.1 and probably 5.0) with power management that will result in > overheating, which will then look like a hardware problem. > > I am currently running continuous buildworlds on stable, and so far > they all succeeded. > > This is what is left to do: > > run buildworld on current or 5.1 until it fails a couple of times > with the below errors, then reboot to 4.8 and run a few buildworlds. > > if they also fail -> hardware problem very likely > if they do not fail -> run again a few buldworlds on current. > if these fail again -> software problem very likely > > could you please also set up this test scenario and report the > outcome? my results will be available some time tomorrow I had exactly these symptoms with my T30 at one point between 5.0 and 5.1. I got random failures in buildworld as well as in a few ports. I never had any failures in normal operations and the errors were, when closely checked, often things like unlinking files and changing file attributes, not in the actual compile. I fixed the problem by building a re-starting the build with -DNOCLEAN. Eventually, the whole thing built. I had to make sure I was not doing something that might leave damaged files around, especially incomplete libraries, that would cause make to not rebuild them. It did take a while. Once I had built the new system and kernel, I installed them and immediately did a complete rebuild of the system and kernel just in case something was not quite right. My T30 has been happily rebuilding 2 or 3 times a week since then with no further problems. Good luck! -- R. Kevin Oberman, Network Engineer Energy Sciences Network (ESnet) Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab) E-mail: oberman@es.net Phone: +1 510 486-8634