122 Commits

Author SHA1 Message Date
Steve Langasek
90e7ee57c3 and the filename was still wrong 2017-10-04 23:50:26 -07:00
Steve Langasek
bc221a8d67 fix a remaining wrong reference to /etc/default/grub.d/40-partuuid 2017-10-04 23:38:22 -07:00
Steve Langasek
a75cbf210b Fix the force-partuuid handling to create the file under
/etc/default/grub.d/ with a .cfg extension as grub expects; it's unclear
how this ever worked without.
2017-10-04 22:51:30 -07:00
Steve Langasek
a5e27120c0 Factor out grub-related diversions and use them consistently, so we
don't end up with wrong os-probe output in our grub.cfg.
2017-10-04 22:33:41 -07:00
Balint Reczey
c5d01a6635 Force partuuid in /etc/default/grub.d/* instead of in /etc/default/grub
This prevents ucf questions upon installation of changed /etc/default/grub
2017-09-27 18:54:09 -04:00
Steve Langasek
3ed1f4fdd1 In a cloud environment, we can rely on the kernel being able to boot the
root filesystem directly, without an initramfs; enable this when
building minimized.
2017-09-26 22:39:09 -04:00
Balint Reczey
45201111a6 Mount using --make-rslave to ensure safe unmounts for rbind mounts 2017-09-25 11:42:28 -04:00
Steve Langasek
801aad2f77 * Improve teardown_mountpoint to recursively find all submounts and
unmount them, instead of working from a hard-coded list.  This makes
  the code resilient against other submounts being added later, including
  downstream.
* Also nuke the sleep / udevadm settle calls in the process, which should
  never be required and slow down the builds.
2017-09-13 13:42:28 -07:00
Steve Langasek
c9ef91e262 Restore the call to 'umount -R $mountpoint' from umount_partition,
accidentally dropped in previous upload.
2017-09-13 12:50:07 -07:00
Robert C Jennings
83b4f5338c live-build/ubuntu-cpc/functions: Add teardown_mountpoint to reverse setup_mountpoint
* live-build/ubuntu-cpc/functions: Add a function, teardown_mountpoint,
  to reverse the work done in setup_mountpoint.  Lack of this function
  has forced users of setup_mountpoint to implement this separately
  and the implementations have diverged. (LP: #1716992)
* live-build/ubuntu-cpc/functions: Remove umount_settle function.
  The was only used where teardown_mountpoint was lacking.
2017-09-13 13:31:12 -05:00
Steve Langasek
a246300940 * live-build/functions: mount tmpfs on /var/cache/apt and var/lib/apt, so
we don't have to leave empty space in our derivative images for packages
  that have been downloaded/installed/removed.  This normally isn't
  relevant for the installed system, since the root filesystem will
  auto-expand in place on the target disk, but lets us ship smaller
  images.
* live-build/functions: also call 'apt-get update' after mounting the
  blank /var/lib/apt.
2017-09-12 14:36:50 -07:00
Colin Watson
6a42148b22 Mount and unmount /dev recursively, to cope with setups where there are
interesting bind-mounts under /dev (e.g. loop devices bind-mounted by
LXD).
2017-09-08 11:48:06 +01:00
Balint Reczey
4d72521491 Fix suppressing kpartx error 2017-08-30 16:46:47 +02:00
Balint Reczey
3b37842219 Use VMDKstream in a more future-proof way 2017-07-24 20:46:51 +02:00
Robert C Jennings
f05a8a238f live-build/functions: Update path for VMDKstream.py
LP: #1706116
2017-07-24 19:20:21 +02:00
Balint Reczey
6c9cf10d72 Depend on pxz and vmdk-stream-converter instead of installing them at runtime 2017-06-14 18:25:15 +02:00
Balint Reczey
aa54a130c3 Depend on packages in main instead of installing them at runtime 2017-06-14 18:25:11 +02:00
Steve Langasek
f75be1c606 Resurrect Balint's fix from
lp:~rbalint/livecd-rootfs/livecd-rootfs-autopkgtest-fix to wrap kpartx
and trap spurious errors, since the problem is still being seen
intermittently on i386 despite us calling sync before kpartx.
2017-04-28 20:44:09 -07:00
Steve Langasek
ca79c05a23 Merge lp:~rbalint/livecd-rootfs/livecd-rootfs-autopkgtest-fix, w/ modifications
Only call sync, don't wrap kpartx and parse its output (doesn't appear to be
needed)
2017-04-25 11:46:26 -07:00
Balint Reczey
21efeb6795 Work around kpartx's nonzero exit due to "Device or resource busy" errors
which still permit kpartx to finally remove loop device (LP: #1684090).
Also sync before calling kpartx to let writing to loop devices finish.
2017-04-21 18:40:19 +02:00
Mathieu Trudel-Lapierre
22ec1d6f6f Sanitize comments. 2017-04-11 23:16:48 -04:00
Mathieu Trudel-Lapierre
46bae4cebf Fix up functions refactoring tfor a proper path to the file. 2017-04-11 17:16:35 -04:00