From owner-freebsd-stable@freebsd.org Thu Sep 14 08:25:46 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 2F128E07BE4 for ; Thu, 14 Sep 2017 08:25:46 +0000 (UTC) (envelope-from borjam@sarenet.es) Received: from cu1176c.smtpx.saremail.com (cu1176c.smtpx.saremail.com [195.16.148.151]) (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 E9BFF6A44E for ; Thu, 14 Sep 2017 08:25:44 +0000 (UTC) (envelope-from borjam@sarenet.es) Received: from [172.16.8.41] (unknown [192.148.167.11]) by proxypop02.sare.net (Postfix) with ESMTPA id 75C8A9DD83A; Thu, 14 Sep 2017 10:25:34 +0200 (CEST) Content-Type: text/plain; charset=utf-8 Mime-Version: 1.0 (Mac OS X Mail 10.3 \(3273\)) Subject: Re: Unusually high "Wired" memory From: Borja Marcos In-Reply-To: Date: Thu, 14 Sep 2017 10:25:33 +0200 Cc: FreeBSD-STABLE Mailing List Content-Transfer-Encoding: quoted-printable Message-Id: References: To: Dan Nelson X-Mailer: Apple Mail (2.3273) 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, 14 Sep 2017 08:25:46 -0000 > On 13 Sep 2017, at 17:56, Dan Nelson via freebsd-stable = wrote: >=20 > 2017-09-12 1:27 GMT-05:00 Borja Marcos : >>=20 >>=20 >>> On 11 Sep 2017, at 11:25, Borja Marcos wrote: >>>=20 >>>> Since I=E2=80=99ve updated a machine to 11.1-STABLE I am seeing a = rather unusual growth of Wired memory. >>>>=20 >>>> Any hints on what might have changed from 11-RELEASE to = 11.1-RELEASE and 11.1-STABLE? >>=20 >> vmstat -z and vmstat -m follow >>=20 >> % vmstat -z >> ITEM SIZE LIMIT USED FREE REQ FAIL = SLEEP >> [...] >> g_bio: 376, 0,15235172, 248,477839025, 0, = 0 >> [...] >=20 >=20 > I think this is the problem - you're leaking g_bio objects (5.7 GB > currently allocated if you multiply SIZE*USED). I'm seeing the same > thing and it looks like it started after the recent ZFS merges from > -current. I have filed a bugreport with some more info ( > https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D222288 ). Confirmed, thank you :) Going back to RELENG-11.1 solved it, so the slip must be quite recent. Borja.