From owner-freebsd-stable@FreeBSD.ORG Fri Jan 18 10:45:53 2008 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 8DAB416A418 for ; Fri, 18 Jan 2008 10:45:53 +0000 (UTC) (envelope-from daniel_k_eriksson@telia.com) Received: from pne-smtpout1-sn1.fre.skanova.net (pne-smtpout1-sn1.fre.skanova.net [81.228.11.98]) by mx1.freebsd.org (Postfix) with ESMTP id 3962813C467 for ; Fri, 18 Jan 2008 10:45:53 +0000 (UTC) (envelope-from daniel_k_eriksson@telia.com) Received: from royal64.emp.zapto.org (195.198.193.104) by pne-smtpout1-sn1.fre.skanova.net (7.3.129) id 478F1C65000584DE; Fri, 18 Jan 2008 10:36:17 +0100 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: quoted-printable Date: Fri, 18 Jan 2008 10:36:30 +0100 Message-ID: <4F9C9299A10AE74E89EA580D14AA10A61A1899@royal64.emp.zapto.org> X-MimeOLE: Produced By Microsoft Exchange V6.5.7235.2 X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: Multiple ZFS pools = panic on shutdown? thread-index: AchZtZ2FuVroRJF+Q62MlP7HaArorA== From: "Daniel Eriksson" To: Cc: pjd@FreeBSD.org Subject: Multiple ZFS pools = panic on shutdown? X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 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, 18 Jan 2008 10:45:53 -0000 Has anyone heard of or experienced any panics ("vput: negative ref cnt" or "vrele: negative ref cnt") when shutting down a box with more than one zpool? I see this on an amd64 RELENG_7_0 (from yesterday) box: syslogd: exiting on signal 15 panic: vrele: negative ref cnt cpuid =3D 0 Uptime: 35m13s Physical memory: 4087 MB Dumping 235 MB: And then it just hangs there. Breaking into DDB does not work. Both pools are constructed from GELI-encrypted disks, so I have a script that runs "geli attach" on all the devices and then it does a "zpool import" on the two pools. On shutdown the script does a "zpool export" on the pools and then it detaches the GELI devices. With only one pool it works just fine. The box in question uses GMIRROR for the system disk (two disks, whole disk mirroring). I will try to recreate this on another machine today. It would be nice to have an easy way to repeat this. /Daniel Eriksson