From owner-freebsd-fs@FreeBSD.ORG Wed Dec 17 20:31:11 2014 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [8.8.178.115]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTPS id 3F9E153E; Wed, 17 Dec 2014 20:31:11 +0000 (UTC) Received: from caida.org (rommie.caida.org [192.172.226.78]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 197BC1767; Wed, 17 Dec 2014 20:31:10 +0000 (UTC) Message-ID: <5491E3CE.2030003@caida.org> Date: Wed, 17 Dec 2014 12:13:02 -0800 From: Daniel Andersen User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.3.0 MIME-Version: 1.0 To: Michael Ranner , freebsd-fs@freebsd.org Subject: Re: Process enters unkillable state and somewhat wedges zfs References: <549152B5.6030100@ranner.eu> In-Reply-To: <549152B5.6030100@ranner.eu> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.18-1 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 17 Dec 2014 20:31:11 -0000 I'm certainly not an expert on this, but from my understanding, I think Andriy Gapon had some thoughts on this particular problem and potential solutions: http://lists.freebsd.org/pipermail/freebsd-fs/2014-November/020482.html At least, I think these problems are related. >From my end, my 'solution' was to stop using nullfs mounts in conjunction with zfs. :( Fortunately, our jail servers don't use zfs, yet. Since I reorganized our data NFS server to stop using nullfs, it's been very stable. ( aside from one crash when it's raid 1 boot drive somehow went offline.. but I'm fairly sure that had nothing whatsoever to do with the ZFS problems. ) Dan On 12/17/2014 01:53 AM, Michael Ranner wrote: > Hello there! > > I have the same problem on FreeBSD 9.3-RELEASE-p6. > > mount_nullfs of ZFS dataset in jail environment. > > php-fpm process get stuck (100% CPU, unkillable) and the specific ZFS dataset is unaccessible. Further processes > accessing the same dataset are hanging in state zfs. > > Other datasets on the same pool are still accessible. > > The stack trace looks very similar to Daniels. > > Is there any progress on this? > > I have many very similar environments (jail, ZFS, nullfs mounts), so I am interested in some solution or workaround. > > -- > Mit freundlichen Grüßen > > Ing. Michael Ranner > > GSM: +43 676 4155044 > Mail: michael@ranner.eu > WWW: http://www.azedo.at/ >