From owner-freebsd-fs@freebsd.org Thu Jan 10 11:08:10 2019 Return-Path: Delivered-To: freebsd-fs@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id DA7941486993 for ; Thu, 10 Jan 2019 11:08:10 +0000 (UTC) (envelope-from borjam@sarenet.es) Received: from cu01176b.smtpx.saremail.com (cu01176b.smtpx.saremail.com [195.16.151.151]) (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 CC6FD768D6 for ; Thu, 10 Jan 2019 11:08:09 +0000 (UTC) (envelope-from borjam@sarenet.es) Received: from [172.16.8.5] (unknown [192.148.167.11]) by proxypop01.sare.net (Postfix) with ESMTPA id DCF059DDCF9; Thu, 10 Jan 2019 12:07:58 +0100 (CET) Content-Type: text/plain; charset=utf-8 Mime-Version: 1.0 (Mac OS X Mail 12.2 \(3445.102.3\)) Subject: MY FAULT. Re: Interesting: ZFS scrub prefetch hurting sequential scrub performance? From: Borja Marcos In-Reply-To: Date: Thu, 10 Jan 2019 12:07:58 +0100 Cc: freebsd-fs@freebsd.org Content-Transfer-Encoding: quoted-printable Message-Id: <77A3F53B-1574-474B-8CDB-ED6A6FECC1C1@sarenet.es> References: <8ECF7513-9DFB-46EF-86BA-DB717D713792@sarenet.es> To: Borja Marcos X-Mailer: Apple Mail (2.3445.102.3) X-Rspamd-Queue-Id: CC6FD768D6 X-Spamd-Bar: / Authentication-Results: mx1.freebsd.org; spf=pass (mx1.freebsd.org: domain of borjam@sarenet.es designates 195.16.151.151 as permitted sender) smtp.mailfrom=borjam@sarenet.es X-Spamd-Result: default: False [-0.18 / 15.00]; ARC_NA(0.00)[]; RCVD_VIA_SMTP_AUTH(0.00)[]; NEURAL_HAM_MEDIUM(-0.80)[-0.802,0]; FROM_HAS_DN(0.00)[]; TO_DN_SOME(0.00)[]; R_SPF_ALLOW(-0.20)[+ip4:195.16.150.0/23]; MV_CASE(0.50)[]; MIME_GOOD(-0.10)[text/plain]; RCVD_TLS_LAST(0.00)[]; DMARC_NA(0.00)[sarenet.es]; NEURAL_SPAM_SHORT(0.19)[0.190,0]; NEURAL_HAM_LONG(-0.77)[-0.766,0]; TO_MATCH_ENVRCPT_SOME(0.00)[]; MX_GOOD(-0.01)[smtp.sarenet.es,smtp.sarenet.es,smtp.sarenet.es]; RCPT_COUNT_TWO(0.00)[2]; RCVD_IN_DNSWL_NONE(0.00)[151.151.16.195.list.dnswl.org : 127.0.10.0]; SUBJECT_ENDS_QUESTION(1.00)[]; FROM_EQ_ENVFROM(0.00)[]; R_DKIM_NA(0.00)[]; MIME_TRACE(0.00)[0:+]; ASN(0.00)[asn:3262, ipnet:195.16.128.0/19, country:ES]; MID_RHS_MATCH_FROM(0.00)[]; IP_SCORE(0.01)[country: ES(0.05)]; RCVD_COUNT_TWO(0.00)[2] X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 10 Jan 2019 11:08:11 -0000 > On 4 Jan 2019, at 11:52, Borja Marcos wrote: >=20 > I have done a test with the old scrub code (vfs.zfs.zfs_scan_legacy=3D1)= and I see a very similar behavior, with the=20 > scrub stalling again. >=20 > Once more, disabling prefetch for the scrub = (vfs.zfs.no_scrub_prefetch=3D1) solves the issue. >=20 > I suffered this problem on 11 at some point but I attributed it = (wrongly!) to hardware problems at the time. >=20 > Not I=E2=80=99ve just found a talk about a new prefetch mechanism for = the scrub by Tom Caputi. Could it be the problem? > https://www.youtube.com/watch?v=3Dupn9tYh917s My apologies for the false alarm! My fault was to use a l2arc in the first place.=20 It seems that the latest updates have made l2arc more detrimental in = relatively low RAM situations. And yes, in those cases in which l2arc is = not recommended in the first place, scrub prefetch can make the = situation worse. But the blame should go to the improper usage of a l2arc, not to the = scrub prefetch instead.=20 Sorry for the confusion and false alarm, although I still think that = this lesson can be included in the guidelines for NOT using a l2arc for = she sake of it. Borja (jumping into a barrel full of tarr and feathers)