From owner-freebsd-stable@freebsd.org Thu Jul 27 18:26:01 2017 Return-Path: Delivered-To: freebsd-stable@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 211D5DAD02A for ; Thu, 27 Jul 2017 18:26:01 +0000 (UTC) (envelope-from mike@sentex.net) Received: from smarthost2.sentex.ca (smarthost2.sentex.ca [IPv6:2607:f3e0:80:80::2]) (using TLSv1 with cipher DHE-RSA-CAMELLIA256-SHA (256/256 bits)) (Client CN "smarthost.sentex.ca", Issuer "smarthost.sentex.ca" (not verified)) by mx1.freebsd.org (Postfix) with ESMTPS id E2401118D for ; Thu, 27 Jul 2017 18:26:00 +0000 (UTC) (envelope-from mike@sentex.net) Received: from lava.sentex.ca (lava.sentex.ca [IPv6:2607:f3e0:0:5::11]) by smarthost2.sentex.ca (8.15.2/8.15.2) with ESMTPS id v6RIPxDq000575 (version=TLSv1 cipher=DHE-RSA-CAMELLIA256-SHA bits=256 verify=NO) for ; Thu, 27 Jul 2017 14:25:59 -0400 (EDT) (envelope-from mike@sentex.net) Received: from [192.168.43.26] (saphire3.sentex.net [192.168.43.26]) by lava.sentex.ca (8.15.2/8.15.2) with ESMTP id v6RIPv9n066547 for ; Thu, 27 Jul 2017 14:25:57 -0400 (EDT) (envelope-from mike@sentex.net) To: FreeBSD-STABLE Mailing List From: Mike Tancsa Subject: rand_harvestq at 100% RELENG_11 Organization: Sentex Communications Message-ID: Date: Thu, 27 Jul 2017 14:25:57 -0400 User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:52.0) Gecko/20100101 Thunderbird/52.2.1 MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 8bit X-Scanned-By: MIMEDefang 2.78 X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 27 Jul 2017 18:26:01 -0000 Not sure if this is normal behaviour or not, but on a recent RELENG_11 r321624, I noticed something odd about rand_harvestq. While doing some disk stress testing of zfs, I was creating a bunch of random files. eg dd if=/dev/urandom > /tmp/junk2 bs=1m count=10000 If I create enough of them, rand_harvestq seems to get pegged at 100% on one of my CPUs last pid: 1014; load averages: 1.36, 0.96, 0.65 up 0+00:17:00 14:22:45 49 processes: 3 running, 45 sleeping, 1 waiting CPU: 0.0% user, 0.0% nice, 24.8% system, 0.0% interrupt, 75.2% idle Mem: 24M Active, 4140K Inact, 12G Wired, 4001M Free ARC: 11G Total, 3144K MFU, 11G MRU, 64K Anon, 19M Header, 1224K Other 11G Compressed, 11G Uncompressed, 1.00:1 Ratio Swap: 40G Total, 40G Free PID USERNAME THR PRI NICE SIZE RES STATE C TIME WCPU COMMAND 11 root 4 155 ki31 0K 64K CPU0 0 55:37 302.56% idle 16 root 1 -16 - 0K 16K CPU3 3 10:39 105.60% rand_harvestq 1014 root 1 20 0 7892K 2668K CPU2 2 0:00 0.03% top And it stays there # ps auxwS | head -5 USER PID %CPU %MEM VSZ RSS TT STAT STARTED TIME COMMAND root 11 295.2 0.0 0 64 - RL 14:05 59:05.25 [idle] root 16 100.0 0.0 0 16 - RL 14:05 11:48.24 [rand_harvestq] root 0 0.0 0.1 0 8400 - DLs 14:05 0:08.96 [kernel] root 1 0.0 0.0 5408 588 - ILs 14:05 0:00.84 /sbin/init Any idea whats going on ? The box is pretty standard CPU: Intel(R) Xeon(R) CPU E3-1220 v3 @ 3.10GHz (3092.90-MHz K8-class CPU) 11.1-STABLE FreeBSD 11.1-STABLE #1 r321624 AMD64 KERNEL is non generic, but it just has a bunch of ethernet drivers removed and pf and pflog added. ---Mike -- -- ------------------- Mike Tancsa, tel +1 519 651 3400 Sentex Communications, mike@sentex.net Providing Internet services since 1994 www.sentex.net Cambridge, Ontario Canada http://www.tancsa.com/