From owner-freebsd-fs@freebsd.org Tue May 17 08:33:28 2016 Return-Path: Delivered-To: freebsd-fs@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 67E1EB3EF9F for ; Tue, 17 May 2016 08:33:28 +0000 (UTC) (envelope-from freebsd-listen@fabiankeil.de) Received: from smtprelay05.ispgateway.de (smtprelay05.ispgateway.de [80.67.31.98]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 2FCAF1F8A for ; Tue, 17 May 2016 08:33:27 +0000 (UTC) (envelope-from freebsd-listen@fabiankeil.de) Received: from [78.35.176.77] (helo=fabiankeil.de) by smtprelay05.ispgateway.de with esmtpsa (TLSv1.2:AES128-GCM-SHA256:128) (Exim 4.84) (envelope-from ) id 1b2aPy-0002dW-TL for freebsd-fs@freebsd.org; Tue, 17 May 2016 10:31:54 +0200 Date: Tue, 17 May 2016 10:27:57 +0200 From: Fabian Keil To: FreeBSD Filesystems Subject: Re: zfs receive stalls whole system Message-ID: <20160517102757.135c1468@fabiankeil.de> In-Reply-To: <0C2233A9-C64A-4773-ABA5-C0BCA0D037F0@ultra-secure.de> References: <0C2233A9-C64A-4773-ABA5-C0BCA0D037F0@ultra-secure.de> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; boundary="Sig_/ZojT=4SLUeXeJZEf2IdOajl"; protocol="application/pgp-signature" X-Df-Sender: Nzc1MDY3 X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.22 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 17 May 2016 08:33:28 -0000 --Sig_/ZojT=4SLUeXeJZEf2IdOajl Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Rainer Duffner wrote: > I have two servers, that were running FreeBSD 10.1-AMD64 for a long time,= one zfs-sending to the other (via zxfer). Both are NFS-servers and MySQL-s= laves, the sender is actively used as NFS-server, the recipient is just a w= arm-standby, in case something serious happens and we don=E2=80=99t want to= wait for a day until the restore is back in place. The MySQL-Slaves are ac= tively used as read-only servers (at the application level, Python=E2=80=99= s SQL-Alchemy does that, apparently). >=20 > They are HP DL380G8 (one CPU, hexacore) with over 128 GB RAM (I think one= has 144, the other has 192). > While they were running 10.1, they used HP P420 RAID-controllers with ind= ividual 12 RAID0 volumes that I pooled into 6-disk RAIDZ2 vdevs. > I use zfsnap to do hourly, daily and weekly snapshots. [...] > Now, when I do a zxfer, sometimes the whole system stalls while the data = is sent over, especially if the delta is large or if something else is read= ing from the disk at the same time (backup agent). >=20 > I had this before, on 10.0 (I believe, we didn=E2=80=99t have this in 9.1= either, IIRC) and it went away in 10.1. Do you use geli for swap device(s)? > It=E2=80=99s very difficult (well, impossible) to debug, because the syst= em totally hangs and doesn=E2=80=99t accept any keypresses. You could try reducing ZFS's deadman timeout to get a panic. On systems with local disks I usually use: vfs.zfs.deadman_enabled: 1 vfs.zfs.deadman_checktime_ms: 5000 vfs.zfs.deadman_synctime_ms: 10000 Fabian --Sig_/ZojT=4SLUeXeJZEf2IdOajl Content-Type: application/pgp-signature Content-Description: OpenPGP digital signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iEYEARECAAYFAlc61g4ACgkQBYqIVf93VJ0shgCaA2wnHQq+AKX3XK7yt5jWKHZ/ rUEAn1IMBjKGvRcA9ZljB/Qy7cY0gLAk =TR3y -----END PGP SIGNATURE----- --Sig_/ZojT=4SLUeXeJZEf2IdOajl--