[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#50223: marked as done (dselect removes diald when it shouldn't)



Your message dated Sat, 27 Nov 1999 17:58:30 -0700 (MST)
with message-id <Pine.LNX.3.96.991127175425.28638I-100000@wakko.deltatee.com>
and subject line delect removes diald when it shouldn't
has caused the attached Bug report to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what I am
talking about this indicates a serious mail system misconfiguration
somewhere.  Please contact me immediately.)

Darren Benham
(administrator, Debian Bugs database)

--------------------------------------
Received: (at submit) by bugs.debian.org; 15 Nov 1999 06:37:48 +0000
Received: (qmail 15644 invoked from network); 15 Nov 1999 06:37:47 -0000
Received: from nautilus.shore.net (207.244.124.104)
  by master.debian.org with SMTP; 15 Nov 1999 06:37:47 -0000
Received: from swift.shore.net (aleph.swift.xxx) [206.243.167.24] 
	by nautilus.shore.net with esmtp (Exim)
	id 11nFlp-0002YV-00; Mon, 15 Nov 1999 01:37:46 -0500
Received: (from swift@localhost)
	by aleph.swift.xxx (8.9.3/8.9.3/Debian 8.9.3-6) id BAA17384;
	Mon, 15 Nov 1999 01:37:10 -0500
Message-Id: <199911150637.BAA17384@aleph.swift.xxx>
From: Matt Swift <swift@alum.mit.edu>
To: Debian Bug Tracking System <submit@bugs.debian.org>
Subject: dselect removes diald when it shouldn't
X-Reportbug-Version: 0.42
X-Mailer: reportbug 0.42
Date: Mon, 15 Nov 1999 01:37:09 -0500

Package: dpkg
Version: 1.4.1.19
Severity: important

I'm using an up-to-date potato system.

The following problem with dselect and the apt method occurs reproducibly.  I
don't have any ideas about how to help debug this any further, so send
suggestions if you need me to try anything.

The diald package currently depends on the fake package `netstd' (this is a
reported bug).  I installed all the dependencies of `netstd' and then removed
`netstd' and some of the dependencies I didn't need on my system.

ACTION/KEYSTROKE                  COMMENT

% dselect                         run dselect
2 <RET> <SPACE>                   go to dselect's selection screen
<RET> <SPACE>                     exit dselect's selection screen; go to
                                    conflict resolution screen:  because diald
	                            is selected, `netstd' and its dependencies,
				    etc. all appear in a suggested
				    configuration.
D                                 choose D)irectly requested state; screen
                                  indicates diald 'installed'; all others on
                                  the screen 'purged' (as they have been)
Q                                 return to dselect main menu via "confirm &
                                  override"
<RET>                             select "Install": I get a message that diald
                                  will be removed!  And indeed it will be, if I
                                  choose to continue...  If I abort and return
                                  to package selection, diald retains the
                                  "installed" status it had before.


But this sequence of keystrokes should NOT cause diald to be removed!

It seems possible to me that this bug could lead to the too-easy removal of a
crucial system package, in which case the severity of this bug should be higher
than `important'.  Diald is happily not crucial, and dselect does announce that
it is about to remove diald.  But I think it is a reasonable to say that the
purpose of the messages about what dselect is about to do is to allow the user
to compare their actions in dselect with their intentions -- NOT to compare
their actions+intentions against bugs in dselect.  When I have high confidence
that what I've requested in dselect is what I meant to request, as is the case
with simple operations, I see no reason to check over the warning messages.

-- System Information
Debian Release: potato
Architecture: i386
Kernel: Linux aleph 2.2.12 #1 Tue Oct 26 20:07:22 EDT 1999 i586

Versions of packages dpkg depends on:
ii  libc6                         2.1.2-10   GNU C Library: Shared libraries an
ii  libncurses4                   4.2-3.4    Shared libraries for terminal hand
ii  libstdc++2.10                 1:2.95.2-3 The GNU stdc++ library            

---------------------------------------
Received: (at 50223-done) by bugs.debian.org; 28 Nov 1999 00:58:33 +0000
Received: (qmail 1193 invoked from network); 28 Nov 1999 00:58:32 -0000
Received: from crash.ab.videon.ca (root@206.75.216.220)
  by master.debian.org with SMTP; 28 Nov 1999 00:58:32 -0000
Received: from wakko.deltatee.com (mail@wakko.powersurfr.com [24.108.53.97])
	by crash.ab.videon.ca (8.9.2/8.9.2) with ESMTP id RAA12266
	for <50223-done@bugs.debian.org>; Sat, 27 Nov 1999 17:58:31 -0700 (MST)
Received: from localhost (wakko.deltatee.com) [127.0.0.1] (jgg)
	by wakko.deltatee.com with smtp (Exim 2.11 #1)
	id 11rsfe-0007XB-00 (Debian); Sat, 27 Nov 1999 17:58:30 -0700
Date: Sat, 27 Nov 1999 17:58:30 -0700 (MST)
From: Jason Gunthorpe <jgg@ualberta.ca>
X-Sender: jgg@wakko.deltatee.com
To: 50223-done@bugs.debian.org
Subject: Re: delect removes diald when it shouldn't
Message-ID: <Pine.LNX.3.96.991127175425.28638I-100000@wakko.deltatee.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII


This is not a bug.

If you use the 'confirm and override' feature of dselect the results will
not be exactly as you have set because APT will refuse to perform an
operation that breaks your system. You have no choice but to remove diald
or place a hold on netstd (or hack the status file and change its
dependencies) until it is properly fixed. 

Basically you selected an impossible situation and told dselect to ignore
the problem and then expected APT to carry out you impossible request,
which it obviosly doesn't do.

Jason


Reply to: