alt installer asks ??? ??? <go back> <continue>

[Carl Karsten]

net/preseeded install, dialog box body: ??? ??? (screen shot attached)

attaching syslog. partman, preseed and a script to wget the installer and run it in a vm.

syslog lines leading up to the ???:

Jun 13 22:18:06 debconf: Adding [ITEMS] -> [The partition tables of the following devices are changed: SCSI1 (0,0,0) (sda) The following partitions are going to be formatted: partition #1 of SCSI1 (0,0,0) (sda) as ext4 partition #5 of SCSI1
Jun 13 22:18:06 debconf: <-- 0
Jun 13 22:18:06 debconf: --> CAPB align
Jun 13 22:18:06 debconf: <-- 0 multiselect backup progresscancel align plugin-detect-keyboard plugin-terminal
Jun 13 22:18:06 debconf: --> INPUT critical partman/confirm_nooverwrite
Jun 13 22:18:06 debconf: <-- 0 question will be asked
Jun 13 22:18:06 debconf: --> GO
Jun 13 22:18:08 debconf: <-- 0 ok
Jun 13 22:18:08 debconf: --> CAPB backup align
Jun 13 22:18:08 debconf: <-- 0 multiselect backup progresscancel align plugin-detect-keyboard plugin-terminal
Jun 13 22:18:08 debconf: --> GET partman/confirm_nooverwrite
Jun 13 22:18:08 debconf: <-- 0 true
Jun 13 22:18:08 debconf: --> RESET partman/confirm_nooverwrite
Jun 13 22:18:08 debconf: <-- 0
Jun 13 22:18:08 debconf: --> SUBST partman/exception_handler TYPE ??? ???
Jun 13 22:18:08 debconf: Adding [TYPE] -> [??? ???]

[Colin Watson (cjwatson)]

parted_server: ../../libparted/arch/linux.c:2958: linux_disk_commit: Assertion `_have_blkpg ()' failed.

[Colin Watson (cjwatson)]

This is due to Linux 3.0's new uname() output.

[Colin Watson (cjwatson)]

Fixed in Debian parted 2.3-6 - I'm just waiting for the Launchpad code import to update so that I can merge that conveniently.

[Launchpad Janitor (janitor)]

This bug was fixed in the package parted - 2.3-6ubuntu1

---------------
parted (2.3-6ubuntu1) oneiric; urgency=low

  * Resynchronise with Debian. Remaining changes:
    - gptsync.patch: On Intel Mac systems, write a synced MBR rather than a
      protective MBR.
    - Add -fno-stack-protector on sparc.
    - loop-partitions.patch: Loop devices can only have one partition, so
      don't generate device names such as "/dev/loop0p1".
    - udevadm-settle.patch: Run 'udevadm settle' either side of rereading
      the partition table if not chrooted, to avoid a variety of races.
    - dmraid.patch: Ensure that device-mapper devices for dmraid arrays do
      not have extra nodes created needlessly, as well as making sure that
      partition nodes for dmraid devices are not probed.
    - fix-head-size-assertion.patch: Fix probe_partition_for_geom() to fail
      gracefully rather than abort with PED_ASSERT().
    - dm_p_separator.patch: Device mapper type should not automatically mean
      add 'p' before the partition number. Fall back to adding it only if
      the previous character is a digit. This complies with kpartx behavior
      and Linux behavior "since the dawn of time".
    - remove-dev_t-dep.patch: Remove incorrect code from arch/linux.c that
      made assumptions about the meaning of dev_t causing mounted partitions
      to not be identified on dmraid devices and extended partition numbers
      >= 16.

parted (2.3-6) unstable; urgency=low

  * Backport from upstream:
    - Accommodate two-component Linux version numbers like 3.0
      (LP: #796865).
  * Rewrite package descriptions and add a Homepage field (thanks, Justin B
    Rye; closes: #615124).
 -- Colin Watson <email address hidden> Thu, 16 Jun 2011 07:36:54 +0100