From owner-freebsd-stable@freebsd.org Fri Apr 12 19:22:28 2019 Return-Path: Delivered-To: freebsd-stable@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 65984158437E for ; Fri, 12 Apr 2019 19:22:28 +0000 (UTC) (envelope-from hausen@punkt.de) Received: from kagate.punkt.de (kagate.punkt.de [217.29.33.131]) (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 1A09D8954A for ; Fri, 12 Apr 2019 19:22:26 +0000 (UTC) (envelope-from hausen@punkt.de) Received: from hugo10.ka.punkt.de (hugo10.ka.punkt.de [217.29.44.10]) by gate2.intern.punkt.de with ESMTP id x3CJMOhh069266; Fri, 12 Apr 2019 21:22:24 +0200 (CEST) Received: from [217.29.46.66] (unassigned [217.29.46.66] (may be forged)) by hugo10.ka.punkt.de (8.14.2/8.14.2) with ESMTP id x3CJMNv2068370; Fri, 12 Apr 2019 21:22:23 +0200 (CEST) (envelope-from hausen@punkt.de) Content-Type: text/plain; charset=utf-8 Mime-Version: 1.0 (Mac OS X Mail 12.4 \(3445.104.8\)) Subject: Re: NVME aborting outstanding i/o and controller resets From: "Patrick M. Hausen" In-Reply-To: Date: Fri, 12 Apr 2019 21:22:23 +0200 Cc: FreeBSD-STABLE Mailing List Content-Transfer-Encoding: quoted-printable Message-Id: <92DAD65A-9BFE-4294-9066-977F498300A3@punkt.de> References: <818CF16A-D71C-47C0-8A1B-35C9D8F68F4E@punkt.de> <58E4FC01-D154-42D4-BA0F-EF9A2C60DBF7@punkt.de> <45D98122-7596-4E8A-8A0D-C33E017C1109@punkt.de> To: Warner Losh X-Mailer: Apple Mail (2.3445.104.8) X-Rspamd-Queue-Id: 1A09D8954A X-Spamd-Bar: -- Authentication-Results: mx1.freebsd.org; spf=pass (mx1.freebsd.org: domain of hausen@punkt.de designates 217.29.33.131 as permitted sender) smtp.mailfrom=hausen@punkt.de X-Spamd-Result: default: False [-2.63 / 15.00]; ARC_NA(0.00)[]; NEURAL_HAM_MEDIUM(-0.99)[-0.993,0]; FROM_HAS_DN(0.00)[]; R_SPF_ALLOW(-0.20)[+ip4:217.29.32.0/20]; MV_CASE(0.50)[]; MIME_GOOD(-0.10)[text/plain]; RCVD_TLS_LAST(0.00)[]; DMARC_NA(0.00)[punkt.de]; NEURAL_HAM_LONG(-1.00)[-0.999,0]; RCVD_COUNT_THREE(0.00)[3]; TO_MATCH_ENVRCPT_SOME(0.00)[]; TO_DN_ALL(0.00)[]; MX_GOOD(-0.01)[cached: mailin.pluspunkthosting.de]; RCPT_COUNT_TWO(0.00)[2]; RCVD_IN_DNSWL_NONE(0.00)[131.33.29.217.list.dnswl.org : 127.0.10.0]; NEURAL_HAM_SHORT(-0.54)[-0.540,0]; IP_SCORE(-0.29)[ipnet: 217.29.32.0/20(-0.79), asn: 16188(-0.64), country: DE(-0.01)]; FROM_EQ_ENVFROM(0.00)[]; R_DKIM_NA(0.00)[]; MIME_TRACE(0.00)[0:+]; ASN(0.00)[asn:16188, ipnet:217.29.32.0/20, country:DE]; MID_RHS_MATCH_FROM(0.00)[] X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 12 Apr 2019 19:22:28 -0000 Hi Warner, thanks for taking the time again =E2=80=A6 > OK. This means that whatever I/O workload we've done has caused the = NVME card to stop responding for 30s, so we reset it. I figured as much ;-) > So it's an intel card. Yes - I already added this info several times. 6 of them, 2.5=E2=80=9C = NVME =E2=80=9Edisk drives=E2=80=9C. > OK. That suggests Intel has a problem with their firmware. I came across this one: https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D211713 Is it more probable that Intel has got buggy firmware here than that =E2=80=9Ewe=E2=80=9C are missing interrupts? The mainboard is the Supermicro H11SSW-NT. Two NVME drive bays share a connector on the mainboard: NVMe Ports ( NVMe 0~7, 10, 11, 14, 15) The H11SSW-iN/NT has tweleve (12) NVMe ports (2 ports per 1 Slim = SAS connector) on the motherboard. These ports provide high-speed, low-latency PCI-E 3.0 x4 = connections directly from the CPU to NVMe Solid State (SSD) drives. This greatly increases SSD data- throughput = performance and significantly reduces PCI-E latency by simplifying driver/software requirements resulting = from direct PCI-E interface from the CPU to the NVMe SSD drives. Is this purely mechanical or do two drives share PCI-E resources? Which = would explain why the problems always come in pairs (nvme6 and nvme7, for example). This afternoon I set up a system with 4 drives and I was not able to = reproduce the problem. (We just got 3 more machines which happened to have 4 drives each and no = M.2 directly on the mainboard). I will change the config to 6 drives like with the two FreeNAS systems = in our data center. > [=E2=80=A6 nda(4) ...] > I doubt that would have any effect. They both throw as much I/O onto = the card as possible in the default config. I found out - yes, just the same. > There's been some minor improvements in -current here. Any chance you = could experimentally try that with this test? You won't get as many I/O = abort errors (since we don't print those), and we have a few more = workarounds for the reset path (though honestly, it's still kinda = stinky). HEAD or RELENG_12, too? Kind regards, Patrick --=20 punkt.de GmbH Internet - Dienstleistungen - Beratung Kaiserallee 13a Tel.: 0721 9109-0 Fax: -100 76133 Karlsruhe info@punkt.de http://punkt.de AG Mannheim 108285 Gf: Juergen Egeling