From owner-freebsd-current@FreeBSD.ORG Thu Jul 4 07:21:25 2013 Return-Path: Delivered-To: current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [8.8.178.115]) by hub.freebsd.org (Postfix) with ESMTP id 06919E48; Thu, 4 Jul 2013 07:21:25 +0000 (UTC) (envelope-from ianf@clue.co.za) Received: from zcs04.jnb1.cloudseed.co.za (zcs04.jnb1.cloudseed.co.za [41.154.0.161]) by mx1.freebsd.org (Postfix) with ESMTP id 9A0DD1272; Thu, 4 Jul 2013 07:21:24 +0000 (UTC) Received: from localhost (localhost [127.0.0.1]) by zcs04.jnb1.cloudseed.co.za (Postfix) with ESMTP id 1E3472A83294; Thu, 4 Jul 2013 09:21:15 +0200 (SAST) X-Virus-Scanned: amavisd-new at zcs04.jnb1.cloudseed.co.za Received: from zcs04.jnb1.cloudseed.co.za ([127.0.0.1]) by localhost (zcs04.jnb1.cloudseed.co.za [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id B6fQopuAYa-P; Thu, 4 Jul 2013 09:21:14 +0200 (SAST) Received: from clue.co.za (unknown [41.154.88.19]) by zcs04.jnb1.cloudseed.co.za (Postfix) with ESMTPSA id 2793B2A83229; Thu, 4 Jul 2013 09:21:14 +0200 (SAST) Received: from localhost ([127.0.0.1] helo=zen.clue.co.za) by clue.co.za with esmtp (Exim 4.80.1 (FreeBSD)) (envelope-from ) id 1Uudqq-0001ka-E7; Thu, 04 Jul 2013 09:21:12 +0200 To: current@freebsd.org Subject: Filesystem wedges caused by r251446 From: "Ian FREISLICH" X-Attribution: BOFH Date: Thu, 04 Jul 2013 09:21:11 +0200 Message-Id: X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.14 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, 04 Jul 2013 07:21:25 -0000 Hi I've been experiencing process wedges writing files. It occurs reliably under "heavy" IO activity, perhaps with large files but I'm not entirely sure what the trigger is. The symptom is that during an installworld, it reliably hangs the install process on /usr/bin/clang (sometimes it hangs earlier at the mtree on /usr) The problem leaves the filesystem in a state where it cannot be cleanly unmounted. A binary search pins the offending commit on r251446. The major differences between the working system and the broken one are, CPU, RAM and arch (i386/amd64) they are otherwise identical Dell R200 servers. Working: CPU: Intel(R) Core(TM)2 Duo CPU E4600 @ 2.40GHz (2400.14-MHz 686-class CPU) RAM: 2G Arch: i386 Broken: CPU: Intel(R) Core(TM)2 Duo CPU E7300 @ 2.66GHz (2666.82-MHz K8-class CPU) RAM: 4G Arch: amd64 The compiler, gcc, clang (trunk or release) has no effect. Working CPU: FreeBSD 10.0-CURRENT #6 r252384: Sun Jun 30 08:36:31 SAST 2013 ianf@fw2:/usr/obj/usr/src/sys/FIREWALL i386 FreeBSD clang version 3.3 (tags/RELEASE_33/final 183502) 20130610 CPU: Intel(R) Core(TM)2 Duo CPU E4600 @ 2.40GHz (2400.14-MHz 686-class CPU) Origin = "GenuineIntel" Id = 0x6fd Family = 0x6 Model = 0xf Stepping = 13 Features=0xbfebfbff Features2=0xe39d AMD Features=0x20100000 AMD Features2=0x1 TSC: P-state invariant, performance statistics real memory = 2147483648 (2048 MB) avail memory = 2076581888 (1980 MB) Broken on: FreeBSD 10.0-CURRENT #19 r251445: Thu Jul 4 08:39:21 SAST 2013 ianf@fw1:/usr/obj/usr/src/sys/FIREWALL amd64 FreeBSD clang version 3.3 (trunk 178860) 20130405 CPU: Intel(R) Core(TM)2 Duo CPU E7300 @ 2.66GHz (2666.82-MHz K8-class CPU) Origin = "GenuineIntel" Id = 0x10676 Family = 0x6 Model = 0x17 Stepping = 6 Features=0xbfebfbff Features2=0x8e39d AMD Features=0x20100800 AMD Features2=0x1 TSC: P-state invariant, performance statistics real memory = 4294967296 (4096 MB) avail memory = 3970727936 (3786 MB) Ian -- Ian Freislich