ubiquity crashed with InstallStepError in configure_hardware()

Bug #536673 reported by Simon Steinbeiß on 2010-03-10
122
This bug affects 20 people
Affects Status Importance Assigned to Milestone
Release Notes for Ubuntu
Undecided
Unassigned
migration-assistant (Ubuntu)
High
Evan
Lucid
High
Evan
partman-base (Ubuntu)
High
Colin Watson
Lucid
High
Colin Watson
ubiquity (Ubuntu)
High
Unassigned
Lucid
High
Unassigned

Bug Description

Release note:

A bug was found in migration-assistant that prevents some users with separate home partitions from completing the install successfully. Users who experience such a crash can disable migration-assistant by first selecting "Try Ubuntu" at the first screen of the installer, then pressing alt-F2 and typing `ubiquity --no-migration-assistant` at the prompt that appears.

Original report follows:
Binary package hint: ubiquity

ubiquity crashed at hardware configuration, most annoyingly this means that grub doesn't get installed properly and leaves you with an unbootable system.

ProblemType: Crash
Architecture: amd64
Date: Wed Mar 10 15:19:14 2010
DistroRelease: Ubuntu 10.04
ExecutablePath: /usr/lib/ubiquity/bin/ubiquity
InterpreterPath: /usr/bin/python2.6
LiveMediaBuild: Xubuntu 10.04 "Lucid Lynx" - Alpha amd64 (20100224.1)
Package: ubiquity 2.1.27
ProcCmdline: /usr/bin/python /usr/lib/ubiquity/bin/ubiquity gtk_ui
ProcEnviron: Error: [Errno 13] Permission denied: '/proc/3712/environ'
ProcVersionSignature: Ubuntu 2.6.32-14.20-generic
PythonArgs: ['/usr/lib/ubiquity/bin/ubiquity', 'gtk_ui']
SourcePackage: ubiquity
Title: ubiquity crashed with InstallStepError in configure_hardware()
Uname: Linux 2.6.32-14-generic x86_64
UserGroups:

Simon Steinbeiß (ochosi) wrote :
tags: removed: need-duplicate-check
Charlie Kravetz (charlie-tca) wrote :

Thanks for reporting this bug and any supporting documentation. Since this bug has enough information provided for a developer to begin work, I'm going to mark it as confirmed and let them handle it from here. Thanks for taking the time to make Ubuntu better!

Changed in ubiquity (Ubuntu):
importance: Undecided → Medium
status: New → Triaged
visibility: private → public
Dion Blundell (dion) wrote :

I have the same issue.
My architecture is different, Intel Atom n450.
I was attempting to install lucid Beta2 netbook-i386

Interestingly enough, I have installed Beta 1, and been running it quite okay since B1 come out.

Could this be a regression?

Should I submit anything else?

Charlie Kravetz (charlie-tca) wrote :

Ubiquity has had many changes since this was reported. Caspar has also been updated as of 2010-04-10 images. Could you test this using the very latest lucid image to see if this has been fixed?

Changed in ubiquity (Ubuntu):
status: Triaged → Incomplete
Dion Blundell (dion) wrote :

I am happy to test. The last .iso I downloaded was beta2 here:
http://www.ubuntu.com/testing
which had the issue.

Is there a daily build of the ISO that I can download somewhere?
If so, I am happy and willing to do that within 24 hours.

Dion

Charlie Kravetz (charlie-tca) wrote :

Thanks for testing this. Daily builds of both the desktop and alternate images are available. For the desktop image: http://cdimages.ubuntu.com/daily-live/current/ , and for the alternate image: http://cdimages.ubuntu.com/daily/current/ . When giving us results, please include the date of the image tested.

I am marking this confirmed since it is still present in the beta2 image, which was dated 2010-04-06.

Changed in ubiquity (Ubuntu):
status: Incomplete → Confirmed
importance: Medium → High
Dion Blundell (dion) wrote :

Hi,
install of Lucid Desktop Live Daily 2004-04-12 fails.

I downloaded the ISO from here: http://cdimages.ubuntu.com/daily-live/20100412/lucid-desktop-i386.iso
I used unetbootin to write it to a USB drive and installed on ASUS EeePC 1005PE

Same crash again.

It throws up an error when trying to copy user data,
it says to check /var/log/syslog
so that is attached.

I'm happy to re-run with anything else that might be useful to you, but would need direction to an FAQ/Howto

I could install Beta1.

I don't know if it is of note, but I have windows and Lucid 10.04 beta1 also installed, my partitions look like this:

My partition table is:
/dev/sda1 31.3GB primary ntfs windows 7
/dev/sda2 32.9MB primary efi bootbooster
/dev/sda3 10.7GB primary fat32 hidden windows recovery partition
/dev/sda4 181GB extended partition
/dev/sda5 2624MB logical linux-swap(v1)
/dev/sda6 157GB logical ext4 home directory for 10.04 beta1, not used for this install
/dev/sda7 10.5GB logical ext4 THIS IS THE PARTITION that I am installing to...
/dev/sda8 10.5GB logical ext4 with a working copy of Lucid 10.04 Beta1 installed from USB
free space 25.3GB

Dion Blundell (dion) wrote :

The date above should read: 2010-04-12 (NOT 2004, sorry)

Dion Blundell (dion) wrote :

I have just re-tested with daily-live
20100416
and the issue persists.

I see the issue has not been assigned to anyone. Does anyone know if this will be fixed before release?
If not it has the potential to wreck-havoc with Asus EeePCs.

Is there a debugging mode for ubiquity so I can see what the issue is?

Dion Blundell (dion) wrote :

Charile,
should this be assigned to Lucid so it is fixed for that release?
Dion

Charlie Kravetz (charlie-tca) wrote :

This is already confirmed, importance high, tagged for lucid. The developers are aware of it.

tags: added: iso-testing
Evan (ev) wrote :

Can anyone that is experiencing this bug please uncomment the "set -x" line in /bin/hw-detect and then attempt to reproduce this bug by running the installer again. When the installer crashes, hit alt-F2 and type `apport-collect 536673`. This will attach detailed logs to this bug report.

Thanks!

Changed in ubiquity (Ubuntu):
status: Confirmed → Incomplete

Architecture: i386
DistroRelease: Ubuntu 10.04
LiveMediaBuild: Ubuntu 10.04 "Lucid Lynx" - Release Candidate i386 (20100419.1)
Package: ubiquity 2.2.20 [modified: bin/hw-detect]
PackageArchitecture: i386
ProcEnviron:
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.32-21.32-generic 2.6.32.11+drm33.2
Tags: lucid
Uname: Linux 2.6.32-21-generic i686
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare

tags: added: apport-collected

apport information

apport information

apport information

apport information

apport information

Jane Atkinson (irihapeti) wrote :

@Evan Dandrea
Logs as requested

Colin Watson (cjwatson) wrote :

Irihapeti, could you try the same thing but run the installer with 'ubiquity -d' from a terminal window? It looks as though the debconf protocol is getting out of sync, and that debug mode should give us a full trace in /var/log/installer/debug.

Evan (ev) wrote :

Irihapeti,

Could I trouble you to run the installer again? Select the "Try Ubuntu" option to bring you to a desktop, then hit alt-f2 and type `ubiquity -d`. This will run the installer in debug mode. When it crashes, type `apport-collect 536673` as you did before.

Thanks!

Architecture: i386
DistroRelease: Ubuntu 10.04
LiveMediaBuild: Ubuntu 10.04 "Lucid Lynx" - Release Candidate i386 (20100419.1)
Package: ubiquity 2.2.20 [modified: bin/hw-detect]
PackageArchitecture: i386
ProcEnviron:
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.32-21.32-generic 2.6.32.11+drm33.2
Tags: lucid
Uname: Linux 2.6.32-21-generic i686
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare

apport information

apport information

apport information

apport information

apport information

Jane Atkinson (irihapeti) wrote :

Here is the info from "ubiquity -d" in terminal.

I enabled "set -x" in /bin/hw-detect as previously. I wasn't quite sure whether to do that or not.

Architecture: i386
DistroRelease: Ubuntu 10.04
Package: ubiquity 2.2.20 [modified: bin/hw-detect]
PackageArchitecture: i386
ProcEnviron:
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.32-21.32-generic 2.6.32.11+drm33.2
Tags: lucid lucid
Uname: Linux 2.6.32-21-generic i686
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare

apport information

apport information

apport information

apport information

apport information

Dion Blundell (dion) wrote :

As requested, I modified hw-detect, to uncomment set -x
I ran
  ubiquity -d
and as you can see above from attached files, ran
  apport-collect 536673
I'm guessing the status can be changed to complete now?

Evan (ev) wrote :

Iihapeti,

Could I trouble you to add 'set -x' to the second line of /usr/lib/ubiquity/migration-assistant/ma-ask and /usr/share/migration-assistant/ma-script-utils and run the installer again, then attach new logs using `apport-collect 536673`?

Thanks a lot and sorry for the run-around. From the logs this appears to be quite a spectacular domino effect.

Architecture: i386
DistroRelease: Ubuntu 10.04
LiveMediaBuild: Ubuntu 10.04 "Lucid Lynx" - Release Candidate i386 (20100419.1)
Package: ubiquity 2.2.20 [modified: usr/lib/ubiquity/migration-assistant/ma-ask usr/share/migration-assistant/ma-script-utils bin/hw-detect]
PackageArchitecture: i386
ProcEnviron:
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.32-21.32-generic 2.6.32.11+drm33.2
Tags: lucid lucid
Uname: Linux 2.6.32-21-generic i686
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare

apport information

apport information

apport information

apport information

apport information

Jane Atkinson (irihapeti) wrote :

I have been getting this error while attempting the install:

"An error occurred while migrating data. More details may be found in /var/log/syslog. The installation will continue, but some or all of the documents and settings you requested may not have been transferred to the installed system."

This happens just before configuring the hardware. I had noticed that I'd not been asked if I wanted to migrate any settings. As I would usually answer "no" to that, it's no big deal for me, but perhaps it should be asking me for this.

I don't mind the "run-around"; it's all in a good cause.

Evan (ev) on 2010-04-23
Changed in ubiquity (Ubuntu Lucid):
status: Incomplete → Triaged
Changed in migration-assistant (Ubuntu Lucid):
importance: Undecided → High
status: New → Triaged
drbovine (brett-isenberg) wrote :

I had been getting the same error at Irihapeti in #43. I turned off the migration assistant during the install (from the command line in a live session ... "$ ubiquity --no-migration-assistant") and it seemed to install fine.

Jane Atkinson (irihapeti) wrote :

Thank you, drbovine.

I tried the same thing, and the installation completed successfully. I was able to boot into it, too.

Dion Blundell (dion) wrote :

Installation completes for me too when I turn migration-assistant off with:
#ubiquity -d --no-migration-assistant

Dion Blundell (dion) wrote :

The problem with this issue is related to _partitioning_

This is a regression from ubuntu 9.10, the issues talked about below do not exist in 9.10

This issue only happens when you are trying to install into the _extended_ partition

The issue is that the partition manager that is part of ubiquity is broken
It may be related to:
https://bugs.launchpad.net/ubuntu/+source/partman-partitioning/+bug/568290

What happens is that Ubuntu 10.04 installer tries to create a logical partition in the extended partition, and this fails if the extended partition exists AND already has 1 or more logical partitions. It succeeds if the extended partition doesn't exist, OR the extended partition is empty.

Ubiquity in 10.04 does something strange with making gaps between the extended partition and the 3 previous primary partitions. It also trys to create gaps between logical partitions.

Basically, partitioning in ubiquity of 10.04 is BROKEN, and that leads to the migration assistant issues.

I have only managed to make it 10.04 install so far because I was telling ubiquity to use/re-format an existing partition. When you choose to use ubiquity to create partitions at install and there already are 4 paritions, eg 3 primary + extended, install fails. The error message is:
"ERROR!!! : Unable to satisfy all constraints on partition [OK]"
"ERROR!!! : Can't have overlapping partitions [OK]"

So, partitioning in the install process of Ubuntu 10.04 is really broken.

Incidentally, migration assistant works IF and ONLY IF you allow ubiquity to create your partitions.

Basically, this issue exists for people with existing operating systems.
So this issue with ubiquity has potential to trash someones system if they already have two operating systems, me for example with windows and a previous ubuntu with a swap file.

I would consider this issue to be significant, and suspect that if this issue of partition manager is released with ubiqity, then ubuntu 10.04 will quickly loose traction as peoples computers are left unbootable, and it is certainly not a viable LTS, as people will be waiting for 10.10 which does install reliably.

I suggest that if it is decided to release this, that you need to go back to the partition manager that came with 9.10

What's more, the ubiquity installer creates a messy partition table.
Picture attached:

Dion Blundell (dion) wrote :

Some more digging, I think this bug is related to this bug here:
https://bugs.launchpad.net/ubuntu/+source/partman-base/+bug/558382

Evan (ev) wrote :

Dion, the bug you describe is entirely different from the one tracked here. This bug is about ubiquity crashing in configure_hardware due to a previous error in migration-assistant. If you're still experiencing your bug in the latest daily-live CDs (http://cdimage.ubuntu.com/daily-live), please file a separate bug.

Evan (ev) wrote :

Can anyone that's experiencing this bug please try with the latest daily-live CD (http://cdimage.ubuntu.com/daily-live/20100426.1/) or any CD in http://cdimage.ubuntu.com/daily-live/ that's newer than 20100426.1.

Let me know via this bug report whether or not the problem is fixed for you. If the crash is still occurring, please put a 'set -x' in the migration-assistant scripts and attach new logs, following the instructions I outlined in comment #36:

https://bugs.edge.launchpad.net/ubuntu/+source/ubiquity/+bug/536673/comments/36

Jane Atkinson (irihapeti) wrote :

The problem appears to be fixed for me. I installed by clicking on the "Install Ubuntu..." icon.

Note that a) I did not resize any partitions (reused an existing one) and b) I did not have any settings to migrate.

Jane Atkinson (irihapeti) wrote :

@Evan:
Update:
I tried again, this time with 20100426.2 version and resizing the partition that I wanted to install to.

The migration assistant gets stuck here:

"Failed to unmount partitions"

"Migration assistant needs to mount a partition but cannot do so because the following mount point could not be unmounted:

/dev/sda1

Please close any applications using these mount points.

Would you like the migration assistant to try to unmount these partitions again?"

If I click "continue", the message just redisplays itself after a split second. I can't get past this point.

I started the install as soon as the disk had booted, so any applications using that partition are not launched by me.

I'm installing to /dev/sdb3. It may be that the previous install I was installing to /dev/sda3, because Lucid is inconsistent about how it labels my hard drives (mix of SATA and IDE).

P.S. I can't quit ubiquity without killing it from a terminal.

Jane Atkinson (irihapeti) wrote :

Note to the foregoing:

I suppose I could quit ubiquity normally if I went back to the previous screen and then pressed "quit". Didn't think of that (short on sleep, I think) :(

Evan (ev) wrote :

Irihapeti,

Ouch, I thought we had this nailed. Can you put 'set -x' in the places I mentioned in comment #36 (https://bugs.edge.launchpad.net/ubuntu/+source/ubiquity/+bug/536673/comments/36), run ubiquity in debug mode (alt-f2 `ubiquity -d` after pressing "Try Ubuntu"), and then do an alt-f2, `apport-collect 536673`.

Also, when the installer says it cannot unmount a device, can you open a terminal and type mount to first figure out where it's mounted, then `sudo lsof /wherever/the/mountpoint/is`. This will let us know what programs are blocking migration-assistant from unmounting that device. Please attach the output of that command to this bug.

Thanks again for all of your help!

Architecture: i386
DistroRelease: Ubuntu 10.04
LiveMediaBuild: Ubuntu 10.04 LTS "Lucid Lynx" - Release i386 (20100426.2)
Package: ubiquity 2.2.23 [modified: usr/lib/ubiquity/migration-assistant/ma-ask usr/share/migration-assistant/ma-script-utils bin/hw-detect]
PackageArchitecture: i386
ProcEnviron:
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.32-21.32-generic 2.6.32.11+drm33.2
Tags: lucid lucid
Uname: Linux 2.6.32-21-generic i686
UserGroups: adm admin cdrom dialout lpadmin plugdev sambashare

apport information

apport information

apport information

apport information

apport information

Jane Atkinson (irihapeti) wrote :

Here is the additional information:

"mount" reports:

ubuntu@ubuntu:~$ mount
aufs on / type aufs (rw)
none on /proc type proc (rw,noexec,nosuid,nodev)
none on /sys type sysfs (rw,noexec,nosuid,nodev)
none on /dev type devtmpfs (rw,mode=0755)
none on /dev/pts type devpts (rw,noexec,nosuid,gid=5,mode=0620)
/dev/sr0 on /cdrom type iso9660 (ro,noatime)
/dev/loop0 on /rofs type squashfs (ro,noatime)
none on /sys/fs/fuse/connections type fusectl (rw)
none on /sys/kernel/debug type debugfs (rw)
none on /sys/kernel/security type securityfs (rw)
none on /dev/shm type tmpfs (rw,nosuid,nodev)
tmpfs on /tmp type tmpfs (rw,nosuid,nodev)
none on /var/run type tmpfs (rw,nosuid,mode=0755)
none on /var/lock type tmpfs (rw,noexec,nosuid,nodev)
none on /lib/init/rw type tmpfs (rw,nosuid,mode=0755)
binfmt_misc on /proc/sys/fs/binfmt_misc type binfmt_misc (rw,noexec,nosuid,nodev)
gvfs-fuse-daemon on /home/ubuntu/.gvfs type fuse.gvfs-fuse-daemon (rw,nosuid,nodev,user=ubuntu)
/dev/sda1 on /mnt/migrationassistant type ext3 (rw)

ubuntu@ubuntu:~$ sudo lsof /mnt/migrationassistant/
lsof: WARNING: can't stat() tmpfs file system /cow
      Output information may be incomplete.
lsof: WARNING: can't stat() fuse.gvfs-fuse-daemon file system /home/ubuntu/.gvfs
      Output information may be incomplete.
ubuntu@ubuntu:~$

The "failed to unmount partitions" message says that the the problem mount point is:

/mnt/migrationassistant/home

When I say I would like the installer to try to unmount these partitions again, it crashed this time.

Here's the lsof output for that directory:

ubuntu@ubuntu:~$ lsof /mnt/migrationassistant/home
lsof: WARNING: can't stat() tmpfs file system /cow
      Output information may be incomplete.
ubuntu@ubuntu:~$

The system installed on /dev/sda1 has a separate /home partition. I wonder if this is part of what's causing the problem?

I'll be unavailable for further testing for about the next five hours or so. (Meetings and things)

Erick Brunzell (lbsolost) wrote :

I have a real dumb question so apologies in advance. Is this occurring while trying to either delete a "casper-rw" partition or trying to perform a manual install with an existing "casper-rw" partition in existence?

If so that can be problematic. I usually reboot with a gparted CD to clean that up if that's the case, if not just shoot me ;^)

Evan (ev) on 2010-04-27
description: updated
Dion Blundell (dion) wrote :

I have tried 20100426.2 and migration assistant either works, or doesn't crash - this is good.
I've tried installing 6 times now, and only once did migration assistant not work, at which time it said "no information available for import" but it _did not crash_, the install was still successful. The other times it detected my Windows 7 install and wanted to import, but I did not allow it to import.

As a side note, it would install successfully under the following conditions:
1) re-using/re-formatting existing partitions
2) starting from scratch partition wise
3) adding partitions to an existing extended partition that had free space

where as previous
scenario 1) would crash
scenario 2) would work
scenario 3) would throw up a partitioning error

For what it's worth, I do not get the errors/issues of post #52 above no matter how hard I try, and can't reproduce them.

Hope this is of use.

Colin Watson (cjwatson) on 2010-04-27
Changed in partman-base (Ubuntu Lucid):
status: New → Fix Committed
importance: Undecided → High
assignee: nobody → Colin Watson (cjwatson)
Changed in migration-assistant (Ubuntu Lucid):
status: Triaged → Fix Committed
assignee: nobody → Evan Dandrea (ev)
Jane Atkinson (irihapeti) wrote :

Just testing 20100427.1 installation. I decided to emulate what a newcomer might do: namely, try to install by double-clicking on the installer icon, find it doesn't work and then try the --no-migration-assistant option.

However, if I do that, migration-assistant is still active and can't be killed. Therefore, the installation still chokes on the "unable to unmount" stage. The only way around it is to reboot. Could that be included in the release notes?

Steve Langasek (vorlon) wrote :

Documented at <https://wiki.ubuntu.com/LucidLynx/ReleaseNotes#Installer%20crash%20with%20a%20separate%20/home%20partition>:

A bug in migration-assistant prevents some users with separate home partitions from completing the install successfully. Users who experience such a crash can disable migration-assistant by selecting "Try Ubuntu" at the first screen of the installer, then pressing Alt-F2 and typing ubiquity --no-migration-assistant at the prompt that appears. (536673)

Changed in ubuntu-release-notes:
status: New → Fix Released
Evan (ev) wrote :

I believe the remaining issues here were fixed with the fix for bug 570765. It would be fantastic if anyone that saw this error to try with the final 10.04 image and report back here if it completed successfully.

Thanks!

Jane Atkinson (irihapeti) wrote :

I've just done an installation using the final image. I proceeded by double-clicking on the "install Ubuntu" icon and chose the same options as previously.

The migration assistant selected no options (maybe it should from my old Hardy install; and I also have a Debian Testing one? However, it's been doing that all along, and as I don't want to migrate anything, I hadn't really paid attention to it - separate issue, anyway).

The installation completed successfully - no problems with unmountable partitions.

Launchpad Janitor (janitor) wrote :

This bug was fixed in the package migration-assistant - 0.6.6

---------------
migration-assistant (0.6.6) lucid-proposed; urgency=low

  * unmount_os can be called without arguments (LP: #536673).
 -- Evan Dandrea <email address hidden> Thu, 29 Apr 2010 09:53:21 +0100

Changed in migration-assistant (Ubuntu Lucid):
status: Fix Committed → Fix Released
Martin Pitt (pitti) wrote :

Copied m-a to -updates since it essentially was included into the final images through ubiquity.

Martin Pitt (pitti) wrote :

ubiquity included this migration-assistant fix and was uploaded straight to lucid-updates.

Is there actually a fix needed for partman-base here?

Changed in ubiquity (Ubuntu Lucid):
status: Triaged → Fix Released
Martin Pitt (pitti) wrote :

Copied ubiquity 2.2.24 from lucid-updates to maverick.

Changed in ubiquity (Ubuntu):
status: Triaged → Fix Released
Steve Langasek (vorlon) wrote :

Based on the available information, I'm backing this out of the release notes.

Martin Pitt (pitti) wrote :

I copied migration-assistant from lucid-proposed to maverick.

Changed in migration-assistant (Ubuntu):
status: Fix Committed → Fix Released
Launchpad Janitor (janitor) wrote :
Download full text (3.3 KiB)

This bug was fixed in the package partman-base - 141ubuntu1

---------------
partman-base (141ubuntu1) maverick; urgency=low

  * Resynchronise with Debian. Remaining changes:
    - Ubiquity integration: If PARTMAN_NO_COMMIT is set, then exit rather
      than running commit.d and finish.d scripts; add a partman-commit
      script; dump extra information to /var/lib/partman/snoop if
      PARTMAN_SNOOP is set; check for per-menu 'no_show_choices' file in
      ask_user and don't reshow the menu if it exists.
    - Don't skip over dmraid devices if the user chooses not to activate
      them.
    - If the only thing mounted on a disk is the installation medium and it
      uses more or less the whole disk, then silently exclude that disk; if
      the installation medium is mounted but doesn't use the whole disk,
      issue a warning that partitioning may be difficult; if anything else
      is mounted, offer to unmount it. partman/filter_mounted=false disables
      this.
    - Use ext4 as the default filesystem for new partitions.
    - Build with -O2 on powerpc to avoid a suspected toolchain bug.
  * Remove cleanup trap in partman-commit, whose only effect is to break
    repeated runs of partman-commit (LP: #536673).

partman-base (141) unstable; urgency=low

  * parted 2.1 changed the semantics of ped_disk_clobber: it now zeroes out
    the first few and last few sectors of the disk, regardless of whether
    there appears to be a valid partition table on it. Unfortunately, this
    means that creating a filesystem on a whole disk device and then calling
    ped_disk_commit_to_dev zeroes the filesystem header we just created. To
    avoid this, call ped_disk_commit_to_dev only if the partition doesn't
    start at sector zero (LP: #539324, #549260).
  * Check return values from asprintf for memory allocation failures.
  * Ignore free space smaller than the grain size of the partition creation
    constraint, rather than only free space smaller than a cylinder.
  * Always leave a small gap at the end of the disk (except on device-mapper
    devices), to avoid confusing mdadm by leaving the MD 0.90 superblock at
    the end of the disk as well as the end of a partition (LP: #527401).
  * Allow preseeding partman/alignment to "cylinder", "minimal", or
    "optimal"; "cylinder" restores old alignment behaviour for the benefit
    of those with crotchety BIOSes, while "optimal" is the default.
  * Don't apply optimal alignment to extended partitions (LP: #558382).

  [ Updated translations ]
  * German (de.po) by Holger Wansing
  * Georgian (ka.po) by Aiet Kolkhi
  * Lithuanian (lt.po) by Kęstutis Biliūnas
  * Tamil (ta.po) by Dr,T,Vasudevan

partman-base (140) unstable; urgency=low

  * Don't warn about data loss on formatted/removed partitions when there
    are no such partitions (LP: #151266).
  * Build against parted 2.2.
  * Use linux-swap(v1) instead of linux-swap(new) to reflect changes to
    parted (thanks, Evan Dandrea; closes: #539908).
  * Apply optimal alignment constraints to new partitions, or when
    maximising an extended partition (LP: #530071).
  * Add an ALIGNMENT_OFFSET command which can be used to detect whether a
...

Read more...

Changed in partman-base (Ubuntu):
status: Fix Committed → Fix Released
Rolf Leggewie (r0lf) wrote :

lucid has seen the end of its life and is no longer receiving any updates. Marking the lucid task for this ticket as "Won't Fix".

Changed in partman-base (Ubuntu Lucid):
status: Fix Committed → Won't Fix
To post a comment you must log in.