From owner-freebsd-fs@FreeBSD.ORG Tue Oct 21 11:50:40 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 5371FB8D; Tue, 21 Oct 2014 11:50:40 +0000 (UTC) Received: from dbh.germany.ru (dbh.germany.ru [188.40.73.20]) (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 D0559F8D; Tue, 21 Oct 2014 11:50:39 +0000 (UTC) Received: from Alex-Varshs-iMac.local (client090-177-185-93.cnt.tvoe.tv [93.185.177.90] (may be forged)) (authenticated bits=0) by dbh.germany.ru (8.14.3/8.14.3) with ESMTP id s9LBoFtJ022535 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NO); Tue, 21 Oct 2014 13:50:26 +0200 (CEST) Message-ID: <544648B4.9030106@putnichek.ru> Date: Tue, 21 Oct 2014 15:51:16 +0400 From: Alex User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:24.0) Gecko/20100101 Thunderbird/24.6.0 MIME-Version: 1.0 To: Andriy Gapon , freebsd-fs@FreeBSD.org Subject: Re: l2_io_error and l2_cksum_bad are not null and growing References: <54454739.1070900@putnichek.ru> <544553CA.1060406@FreeBSD.org> In-Reply-To: <544553CA.1060406@FreeBSD.org> Content-Type: text/plain; charset=UTF-8; format=flowed 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: Tue, 21 Oct 2014 11:50:40 -0000 Hello. As I see this patch seems to be pretty old. How can it be that it haven't been included in the stable branche since then? The problem is that I can't afford making experiments on the production server. Is there any official solution of the issue? Alex 10/20/14 10:26 PM, Andriy Gapon пишет: > On 20/10/2014 20:32, Alex wrote: >> Hello. >> >> We seem to have a problem with l2arc on zfs system on 10.1-BETA1 FreeBSD >> 10.1-BETA1 #1 r271710 server. The server has a l2arc cache configured from Intel >> SSD 480G disk. The problem is that l2_io_error and l2_cksum_bad values are >> constantly growing, however there are no traces of any hardware malfunctioning. >> As for now, the values are >> kstat.zfs.misc.arcstats.l2_io_error: 1501 >> kstat.zfs.misc.arcstats.l2_cksum_bad: 19480 > Please see if the following patch might help > https://github.com/avg-I/freebsd/compare/review/l2arc-write-target-size.diff > >> Here is the output of zpool status: >> pool: zpool >> state: ONLINE >> scan: none requested >> config: >> >> NAME STATE READ WRITE CKSUM >> zpool ONLINE 0 0 0 >> mirror-0 ONLINE 0 0 0 >> diskid/DISK-WD-WMC1P0DFSF47p2 ONLINE 0 0 0 >> diskid/DISK-WD-WMC1P0DEFERYp2 ONLINE 0 0 0 >> logs >> gpt/zil0 ONLINE 0 0 0 >> cache >> gpt/cache0 ONLINE 0 0 0 >> >> errors: No known data errors >> >> Here is the output of zfs-stats -L: >> ------------------------------------------------------------------------ >> ZFS Subsystem Report Mon Oct 20 19:29:38 2014 >> ------------------------------------------------------------------------ >> >> L2 ARC Summary: (DEGRADED) >> Passed Headroom: 72.42m >> Tried Lock Failures: 360.55m >> IO In Progress: 65 >> Low Memory Aborts: 101 >> Free on Write: 7.09k >> Writes While Full: 16.90k >> R/W Clashes: 11 >> Bad Checksums: 19.48k >> IO Errors: 1.50k >> SPA Mismatch: 1.18m >> >> L2 ARC Size: (Adaptive) 555.88 GiB >> Header Size: 0.21% 1.17 GiB >> >> L2 ARC Evicts: >> Lock Retries: 10 >> Upon Reading: 0 >> >> L2 ARC Breakdown: 272.76m >> Hit Ratio: 0.35% 949.30k >> Miss Ratio: 99.65% 271.81m >> Feeds: 2.19m >> >> L2 ARC Buffer: >> Bytes Scanned: 20.45 PiB >> Buffer Iterations: 2.19m >> List Iterations: 139.97m >> NULL List Iterations: 5.58k >> >> L2 ARC Writes: >> Writes Sent: 100.00% 567.37k >> >> Any help is welcome. Best regards >> >