[01:34]  * hardwire stretches
[01:34]  * hardwire pops neck
[01:34]  * hardwire presses the red button
[01:35] <hardwire> I love remote ubuntu installs
[08:43] <cjwatson> foka: merged, thanks!
[11:05] <CIA-59> usb-creator: evand * r11 usb-creator/ (5 files in 4 dirs): (log message trimmed)
[11:05] <CIA-59> usb-creator: * Replaced the log-output backed command wrapper with a simpler solution
[11:05] <CIA-59> usb-creator:  (LP: #269044).
[11:05] <CIA-59> usb-creator: * Now logging to ~/.usb-creator.log instead of just stdout.
[11:05] <CIA-59> usb-creator: * Moved the persistence code from the backend to scripts/install.py and
[11:05] <CIA-59> usb-creator:  wired up the persistence scale in the frontend.
[11:05] <CIA-59> usb-creator: * The backend now reports the task description in addition to the progress
[13:13] <CIA-59> clock-setup: cjwatson * r197 clock-setup/debian/ (changelog clock-setup.postinst):
[13:13] <CIA-59> clock-setup: * Fix handling of progress bar cancellation (LP: #262500):
[13:13] <CIA-59> clock-setup:  - db_progress INFO might be cancelled if the user is quick; cope with
[13:13] <CIA-59> clock-setup:  that.
[13:13] <CIA-59> clock-setup:  - Reset the progresscancel capability when we're finished.
[13:15] <CIA-59> clock-setup: cjwatson * r198 clock-setup/debian/changelog: releasing version 0.96ubuntu2
[18:29] <CIA-59> ubiquity: cjwatson * r2815 ubiquity/ (debian/changelog gui/glade/stepKeyboardConf.glade):
[18:29] <CIA-59> ubiquity: Pressing Enter in the keyboard test box activates the Forward button
[18:29] <CIA-59> ubiquity: (LP: #225548).
[20:09] <foka> cjwatson, Thanks!
[20:23] <xivulon> cjwatson, evand re #269946
[20:24] <xivulon> I have checked and that does not affect nested filesystem
[20:24] <xivulon> I.E. I can use open+O_Direct to open a file inside ext3 inside ntfs
[20:25] <xivulon> hence either a workaround is implemented for O_DIRECT in the kernel/ntfs-3g
[20:26] <xivulon> or we should try to avoid O_DIRECT calls in partman and the installer.
[20:29] <xivulon> Only calls that manipulate the virtual disks directly are relevant, once the virtual disks are loop mounted it should be ok.
[20:45] <xivulon> calls that hit /boot should also be checked (although the plan in this case might be to keep a local /boot and rsync it to the host on update-grub because of vfat limitations)
[20:47] <xivulon> I will have a word with cking and szaka to see if a workaround can be implemented easily in ntfs first.
[23:17] <CIA-59> ubiquity: cjwatson * r2816 ubiquity/ (debian/changelog ubiquity/components/console_setup.py):
[23:17] <CIA-59> ubiquity: Improve handling of default keymap options used when setting the X
[23:17] <CIA-59> ubiquity: keymap on the fly. In particular, set the lv3:ralt_switch option for
[23:17] <CIA-59> ubiquity: non-US layouts so that AltGr can be tested (LP: #258153).
[23:18] <CIA-59> ubiquity: cjwatson * r2817 ubiquity/ (debian/changelog ubiquity/components/console_setup.py): Clear out previous XKB options when setting a new X keymap.