From owner-freebsd-fs@FreeBSD.ORG Mon Jun 13 19:23:48 2011 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id B5C5D106566C for ; Mon, 13 Jun 2011 19:23:47 +0000 (UTC) (envelope-from mirror176@hotmail.com) Received: from snt0-omc4-s30.snt0.hotmail.com (snt0-omc4-s30.snt0.hotmail.com [65.55.90.233]) by mx1.freebsd.org (Postfix) with ESMTP id 8BF808FC13 for ; Mon, 13 Jun 2011 19:23:47 +0000 (UTC) Received: from SNT105-W27 ([65.55.90.200]) by snt0-omc4-s30.snt0.hotmail.com with Microsoft SMTPSVC(6.0.3790.4675); Mon, 13 Jun 2011 12:23:46 -0700 Message-ID: X-Originating-IP: [24.56.42.84] From: Edward Sutton To: Date: Mon, 13 Jun 2011 12:23:46 -0700 Importance: Normal MIME-Version: 1.0 X-OriginalArrivalTime: 13 Jun 2011 19:23:46.0901 (UTC) FILETIME=[6A9D8850:01CC29FF] Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable X-Content-Filtered-By: Mailman/MimeDel 2.1.5 Subject: Re: zfs mirror: 1 disk lost, corrupted other disk. crashes zfs tools and panics system X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 13 Jun 2011 19:23:48 -0000 Wanted to report from a state I could describe without confusion of steps= I had taken. Restored DD copy and ran `zpool import -fFX zroot` from the s= ame -current disk mentioned earlier and this time it did roll back a few mi= nutes and gave access to the corrupt filesystems. I must have touched it mo= re with an older copy every time though I thought that was not the first ti= me I started with -current. -X still is not defined in the manpage but is r= eferenced a lot online. I still have the disk copy that can be restored in the corrupt form if an= yone is interested in fixing issues such as scrubbing without the v28 rollb= ack taking place. The scrub was when it got stuck in a panic on import beca= use the scrub would resume and would cause a panic in doing so. Current was prone to occasional random crashes on my system so it took a = while to copy off the data in a successful run=2C though that may be my har= dware (which seems most stable after being up and running for a few hours u= nder full cpu load such as boinc). =