From owner-freebsd-stable@FreeBSD.ORG Tue Aug 27 15:32:06 2013 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) (using TLSv1 with cipher ADH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTP id AC54BF1B for ; Tue, 27 Aug 2013 15:32:06 +0000 (UTC) (envelope-from danny@cs.huji.ac.il) Received: from kabab.cs.huji.ac.il (kabab.cs.huji.ac.il [132.65.16.84]) (using TLSv1 with cipher ECDHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.freebsd.org (Postfix) with ESMTPS id 5D9BF2F4E for ; Tue, 27 Aug 2013 15:32:06 +0000 (UTC) Received: from pampa.cs.huji.ac.il ([132.65.80.32]) by kabab.cs.huji.ac.il with esmtp id 1VELFR-000J7V-Hv; Tue, 27 Aug 2013 18:32:01 +0300 X-Mailer: exmh version 2.7.2 01/07/2005 with nmh-1.3 To: Konstantin Belousov Subject: Re: another? NFS deadlock on 9.2-PRERELEASE In-reply-to: <20130827140412.GQ4972@kib.kiev.ua> References: <1740270726.14074212.1377605362382.JavaMail.root@uoguelph.ca> <20130827140412.GQ4972@kib.kiev.ua> Comments: In-reply-to Konstantin Belousov message dated "Tue, 27 Aug 2013 17:04:12 +0300." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 27 Aug 2013 18:32:01 +0300 From: Daniel Braniss Message-ID: Cc: Rick Macklem , freebsd-stable X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 27 Aug 2013 15:32:06 -0000 > > --wdMRLhhF94AmkTAJ > Content-Type: text/plain; charset=us-ascii > Content-Disposition: inline > Content-Transfer-Encoding: quoted-printable > > On Tue, Aug 27, 2013 at 05:00:14PM +0300, Daniel Braniss wrote: > > > Daniel Braniss wrote: > > > > > Daniel Braniss wrote: > > > > > > I upgraded our web server, and only after 3 hours it hung :-( > > > > > > (as a side note, I have 2 other web servers, also running 9.2 > > > > > > doing > > > > > > great :-) > > > > > > go figure. > > > > > >=20 > > > > > > anyways, in > > > > > > ftp://ftp.cs.huji.ac.il/users/danny/freebsd/core.txt/0 > > > > > >=20 > > > > > > is the info after a forced panic. > > > > > >=20 > > > > > Looks like the same hang to me. Several threads are sleeping on > > > > > "pgrbwt" > > > > > and lots are waiting for an NFS vnode lock. > > > > >=20 > > > > > It should be fixed in RC3 (or revert r250907). If it still hangs > > > > > with > > > > > RC3 (or r250907 reverted), email again. > > > > >=20 > > > > im following stable, hence it's till calling itself 9.2-PRERELEASE, > > > > but > > > > I did a sync this morning - local time, after rc3 was anounced. > > > > but after 3.45 minutes is hung, data in > > > > ftp://ftp.cs.huji.ac.il/users/danny/freebsd/core.txt/1 > > > >=20 > > > > I can't easely revert r250907, since i'm using mercuriall, but if > > > > someone > > > > can send me the pre r250907 files, i'll try. > > > >=20 > > > r254947, which was committed to stable/9 a few hours ago is believed to > > > fix the problem. Please update your stable/9 to post-r254947 and try it. > > >=20 > > the current kernel has that fix (sys/kern/uipc_syscalls.c) > > and if you check the core.txt/1 you will see no pgrbwt, only newnsf ... > > There is almost no useful information in the core.txt/1. > Provide the known data for the deadlock. maybe the word deadlock is too strong, the host is diskless (one of many) and so when NFS stops working/respondig, it hangs. I will now make it dataless - the / will be from local disk, see if that makes it easier to debug.