Cannot install with a blank password

Bug #280014 reported by Emmet Hikory
10
This bug affects 2 people
Affects Status Importance Assigned to Milestone
installation-guide (Debian)
Fix Released
Unknown
installation-guide (Ubuntu)
Won't Fix
Undecided
Emmet Hikory
ubiquity (Ubuntu)
Fix Released
Undecided
Emmet Hikory
user-setup (Debian)
Fix Released
Unknown
user-setup (Ubuntu)
Fix Released
Undecided
Emmet Hikory

Bug Description

Binary package hint: user-setup

When performing an install, the user cannot proceed without entering a password. This is the case both interactively and when a blank password has specifically been seeded for noninteractive installation.

As in the normal case of events, the user will not want to install with a blank password, it makes sense to have a check that the password contains a value. In order to support installation with a blank password, there should be a preseed value to specifically allow this use case.

Emmet Hikory (persia)
Changed in user-setup:
assignee: nobody → persia
status: New → In Progress
Changed in ubiquity:
assignee: nobody → persia
status: New → In Progress
Revision history for this message
Emmet Hikory (persia) wrote :

Solution proposed in the attached branches is to ass an additional passwd/allow-password-empty boolean preseed value, and to have user-setup check this before presenting the user-setup/password-empty error.

The password empty check has been removed from ubiquity. In the event that a user proceeds with an empty password, and it is not permitted, the user-setup/password-empty error will return the user to the ubiquity interface for input of the password.

Revision history for this message
Colin Watson (cjwatson) wrote :

user-setup change applied, though I merged it too early to include the bug number. (I've remerged the branch so that it's there for archaeological purposes.)

user-setup (1.20ubuntu8) intrepid; urgency=low

  * Add support and template for passwd/allow-password-empty as a
    boolean preseed value that overrides the test for an empty
    password.

 -- Emmet Hikory <email address hidden> Tue, 07 Oct 2008 18:23:28 +0900

Changed in user-setup:
status: In Progress → Fix Released
Revision history for this message
Colin Watson (cjwatson) wrote :

With regard to your ubiquity change, I think I'd rather have ubiquity check passwd/allow-password-empty and continue to disable the next button for empty password/verified-password entry boxes if it's false; I don't think this should be too difficult. My reasoning is:

  * If you don't know about the allow-password-empty change, it's a bit of a weird UI change.
  * It's better to handle problems as early as possible rather than having to go through user-setup and back. Out of necessity, there are some problems that we can only deal with by the latter method, but where it's feasible we should prefer to simply prevent the user from going forward until they've filled in the necessary information.

Emmet Hikory (persia)
Changed in installation-guide:
assignee: nobody → persia
status: New → In Progress
Revision history for this message
Emmet Hikory (persia) wrote :

I've updated the ubiquity branch with a change that preserves the previous user interaction except in cases where passwd/allow-password-empty is set to true, in which case the Forward button is enabled while the password entry is blank. I was unable to expose user-setup/password-empty in the ubiquity UI for any of the permutations I tested with this updated branch.

Colin Watson (cjwatson)
Changed in ubiquity:
status: In Progress → Fix Committed
Changed in installation-guide:
status: Unknown → New
Changed in user-setup:
status: Unknown → New
Revision history for this message
Emmet Hikory (persia) wrote :

The applied fix to user-setup was quite effectively shot down as being even sensible by Frans Pop in BTS 501849. While I'd personally be happy not to revert for Intrepid given the timeframe, I'll plan on unimplementing this and preparing different functionality for Jaunty to meet the target goals.

If there is any desire to unimplement for Intrepid, please comment to this bug, and I'll address the root issue in another manner.

Revision history for this message
Colin Watson (cjwatson) wrote :

Since Frans hadn't considered the case of embedded devices without keyboard input, I have no problem with this remaining in place for Intrepid.

Revision history for this message
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package ubiquity - 1.10.5

---------------
ubiquity (1.10.5) intrepid; urgency=low

  [ Evan Dandrea ]
  * Filter out the net subsystem when calling update-dev to prevent the
    network connection from resetting (LP: #276383).
  * If ubiquity is installing from a disk, such as a USB drive, then default
    to writing grub to the disk that /boot or / is to be written to, as (hd0)
    will most likely be the installation medium (LP: #282037).
  * Automatic update of included source packages: apt-setup
    1:0.37ubuntu5, partman-auto 78ubuntu3, partman-base 121ubuntu7,
    partman-target 55ubuntu4, user-setup 1.20ubuntu9.

  [ Colin Watson ]
  * Fix typo in architecture detection for ntfsprogs dependency.
  * Work around lpia having DEB_HOST_ARCH_CPU=i686 (!).
  * Disable window minimise buttons if the installer is running in
    standalone mode (LP: #249045).
  * Update imported translations from gtk+2.0 2.14.3-0ubuntu3.
  * Update translations from Launchpad (LP: #144741, #218636, #277451).

  [ Emmet Hikory ]
  * Honor preseeded passwd/allow-password-empty (LP: #280014)

 -- Evan Dandrea <email address hidden> Mon, 13 Oct 2008 01:34:45 -0400

Changed in ubiquity:
status: Fix Committed → Fix Released
Revision history for this message
Colin Watson (cjwatson) wrote :

For the record, I renamed this to user-setup/allow-password-empty for consistency with user-setup/password-empty (passwd/* really ought to be renamed to user-setup/* in general; the passwd/* names are historical). I'll keep compatibility with the old name for one release.

Emmet Hikory (persia)
Changed in installation-guide:
status: In Progress → Won't Fix
Revision history for this message
Emmet Hikory (persia) wrote :

I'm marking the installation-guide change won't fix, have closed the Debian bugs (or attempted to do so), and intend to revert this as part of bug #347207. There are other ways to achieve the goal (which isn't a very good goal anyway), and it's easier to stay in line with trunk.

Changed in user-setup (Debian):
status: New → Fix Released
Changed in installation-guide (Debian):
status: New → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.