NetBSD Problem Report #42322
From www@NetBSD.org Sun Nov 15 11:53:55 2009
Return-Path: <www@NetBSD.org>
Received: from mail.netbsd.org (mail.netbsd.org [204.152.190.11])
by www.NetBSD.org (Postfix) with ESMTP id 29B3E63B8CD
for <gnats-bugs@gnats.netbsd.org>; Sun, 15 Nov 2009 11:53:55 +0000 (UTC)
Message-Id: <20091115115354.EE99063B844@www.NetBSD.org>
Date: Sun, 15 Nov 2009 11:53:54 +0000 (UTC)
From: mr.qweo@gmail.com
Reply-To: mr.qweo@gmail.com
To: gnats-bugs@NetBSD.org
Subject: ext2fs support incomplete enough for a core dump or two
X-Send-Pr-Version: www-1.0
>Number: 42322
>Category: bin
>Synopsis: ext2fs support incomplete enough for a core dump or two
>Confidential: no
>Severity: serious
>Priority: medium
>Responsible: bin-bug-people
>State: open
>Class: sw-bug
>Submitter-Id: net
>Arrival-Date: Sun Nov 15 11:55:00 +0000 2009
>Last-Modified: Mon Nov 23 10:15:02 +0000 2009
>Originator: rider
>Release: 5.0-STABLE
>Organization:
>Environment:
NetBSD stallion ...
>Description:
Apparently, NetBSD support of the Second Extended FS (heirs) is partial,
just as Linux support of Berkeley FFS (heirs), and act of running fsck_ext2fs (8) on a journaled ext2fs counterpart may be considered imprudent,
but since there's no BUGS section there in fsck_ext2fs(8) I did it.
Consequently I've got a nice core dump (no, I can't really assist with audit -
I (i. e. me, not NetBSD ;-) ) nuked the poor thing then (BTW, would a wizard
with enough cross-platform experience suggest an FS sound for a home partition?)
Now don't you think that this is the end? :-)
I did the necessary maintance using native tools, re-booted back into NetBSD['s
rescue system] and executed vi(1) to edit a configuration file on
another such file system. Guess what? :-)
>How-To-Repeat:
Create a journaled ext2fs, make it dirty and run fsck_ext2fs(8) on it.
If you're feeling adventurous (C) the INSTALL document, go for vi(1).
>Fix:
Using native tools and file systems may be considered a work-around :-)
>Audit-Trail:
From: rider <mr.qweo@gmail.com>
To: gnats-bugs@netbsd.org
Cc:
Subject: Re: bin/42322: ext2fs support incomplete enough for a core dump or
two
Date: Sat, 21 Nov 2009 10:17:00 -0200
Well, I tried to follow the INSTALL document advice not to abuse
attachments when sending mail to NetBSD mailing lists, but since
no one contacted me so far, I'd send the core dumps, sorry for any
inconvenience.
From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
To: gnats-bugs@NetBSD.org
Cc: gnats-admin@NetBSD.org, netbsd-bugs@NetBSD.org, mr.qweo@gmail.com,
tsutsui@ceres.dti.ne.jp
Subject: Re: bin/42322: ext2fs support incomplete enough for a core dump or
two
Date: Sat, 21 Nov 2009 21:30:08 +0900
>Synopsis: ext2fs support incomplete enough for a core dump or two
You mean, the synopsis should be
"fsck_ext2fs(8) dumps core on checking journaled ext2fs" ?
Then, it might be worth to note actuall command how you did create
your journaled ext2fs (by mke2fs with what option etc.) in
How-To-Repeat section so that people can reproduce and debug your problem.
---
Izumi Tsutsui
From: Izumi Tsutsui <tsutsui@ceres.dti.ne.jp>
To: mr.qweo@gmail.com
Cc: gnats-bugs@NetBSD.org, gnats-admin@NetBSD.org, netbsd-bugs@NetBSD.org,
tsutsui@ceres.dti.ne.jp
Subject: Re: bin/42322: ext2fs support incomplete enough for a core dump or
two
Date: Sat, 21 Nov 2009 22:17:25 +0900
> > On 21/11/2009, Izumi Tsutsui <tsutsui@ceres.dti.ne.jp> wrote:
> > You mean, the synopsis should be
> > "fsck_ext2fs(8) dumps core on checking journaled ext2fs" ?
> if you're talking about the bug report,
> I've got multiple core dumps (from fsck_ext2fs(8), fdisk(8) and
> vi(1) so far).
Well, it's better to mention what each actual
problem you get on what command per each command,
i.e. what command you typed and what result what you got
rather than personal impressions in a problem report.
NetBSD doesn't support ext2fs journaling
and fsck_ext2fs(8) man page should clarify it,
but commands should not be dump core anyway.
---
Izumi Tsutsui
From: rider <mr.qweo@gmail.com>
To: gnats-bugs@netbsd.org
Cc: gnats-admin@netbsd.org, netbsd-bugs@netbsd.org
Subject: Re: bin/42322: ext2fs support incomplete enough for a core dump or
two
Date: Sat, 21 Nov 2009 11:00:10 -0200
On 21/11/2009, Izumi Tsutsui <tsutsui@ceres.dti.ne.jp> wrote: >
You mean, the synopsis should be > "fsck_ext2fs(8) dumps core on
checking journaled ext2fs" ? There's BUGS section after all if you
mean manual page fix, and if you're talking about the bug report,
I've got multiple core dumps (from fsck_ext2fs(8), fdisk(8) and
vi(1) so far).
> Then, it might be worth to note actuall command how you did
create > your journaled ext2fs (by mke2fs with what option etc.)
in > How-To-Repeat section so that people can reproduce and debug
your problem. IIRC I used no options besides -L and -m when creating
file system, then I tuned it so no known to be problematic features
like B-Tree directory indexing would be enabled.
From: rider <mr.qweo@gmail.com>
To: gnats-bugs@netbsd.org
Cc: gnats-admin@netbsd.org, netbsd-bugs@netbsd.org
Subject: Re: bin/42322: ext2fs support incomplete enough for a core dump or
two
Date: Sat, 21 Nov 2009 11:34:27 -0200
On 21/11/2009, Izumi Tsutsui <tsutsui@ceres.dti.ne.jp> wrote:
>
> Well, it's better to mention what each actual
> problem you get on what command per each command,
> i.e. what command you typed and what result what you got
> rather than personal impressions in a problem report.
I'd rather upload the dumps if there's some one inerested
enough to host it (I've got handful of technical difficulties
now, so I can't really host these, nor I can mail them.)
From: David Holland <dholland-bugs@netbsd.org>
To: gnats-bugs@NetBSD.org
Cc: gnats-admin@netbsd.org, netbsd-bugs@netbsd.org, mr.qweo@gmail.com
Subject: Re: bin/42322: ext2fs support incomplete enough for a core dump or
two
Date: Mon, 23 Nov 2009 07:57:08 +0000
On Sat, Nov 21, 2009 at 01:35:02PM +0000, rider wrote:
> > Well, it's better to mention what each actual
> > problem you get on what command per each command,
> > i.e. what command you typed and what result what you got
> > rather than personal impressions in a problem report.
>
> I'd rather upload the dumps if there's some one inerested
> enough to host it (I've got handful of technical difficulties
> now, so I can't really host these, nor I can mail them.)
Core dumps themselves are not very useful without the exact program
image that generated them, and even then they still aren't very useful
unless the program was compiled with debug information.
It is in general a lot faster and more productive to report what you
did and what happened as a result.
--
David A. Holland
dholland@netbsd.org
From: rider <mr.qweo@gmail.com>
To: David Holland <dholland-bugs@netbsd.org>
Cc: gnats-bugs@netbsd.org, gnats-admin@netbsd.org, netbsd-bugs@netbsd.org
Subject: Re: bin/42322: ext2fs support incomplete enough for a core dump or
two
Date: Mon, 23 Nov 2009 08:14:04 -0200
On 23/11/2009, David Holland <dholland-bugs@netbsd.org> wrote: >
Core dumps themselves are not very useful without the exact program
> image that generated them, and even then they still aren't very
useful > unless the program was compiled with debug information.
The programs involved were these shipped with NetBSD 5.0.1 release
back then. > > It is in general a lot faster and more productive
to report what you > did and what happened as a result. Well, I've
got a dirty journaled Second Extended FS, then executed fsck_ext2fs(8)
on it (without any options IIRC) and it dumped core. Then I executed
vi(1) on another journaled SE FS and *it* dumped core :-) Then
still, I executed fdisk(8) on an MBR-labeled disk, modified the
partition table, and on writing, the thing dumped core too.
(Contact us)
$NetBSD: query-full-pr,v 1.39 2013/11/01 18:47:49 spz Exp $
$NetBSD: gnats_config.sh,v 1.8 2006/05/07 09:23:38 tsutsui Exp $
Copyright © 1994-2007
The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.