simplify adt-virt-lxc startup checks

So much easier! http://anonscm.debian.org/gitweb/?p=autopkgtest/autopkgtest.git;a=commitdiff;h=e7a0f5b31

This bug was fixed in the package autopkgtest - 2.7

---------------
autopkgtest (2.7) unstable; urgency=medium

  * adt-virt-lxc: Drop all the file-based checks for container bootup, and
    replace with "lxc-attach runlevel" to wait until the container booted to a
    numeric runlevel. This greatly simplifies the logic and drops the
    cloud-init requirement so that this runner is now suitable for Debian or
    debootstrap-based Ubuntu containers. (LP: #1273725)
  * adt-virt-lxc.1: Drop --gain-root in notes/example. It's unnecessary
    (adt-run will install fakeroot if missing) and confusing (as it's not
    related to LXC specifically and otherwise not really used/useful).
  * adt-virt-lxc.1: Add example how to create a Debian container, and some
    cleanup of obsolete commentary.
  * adt-virt-lxc: Add -s/--sudo option to run lxc-* commands through sudo; use
    this option if you run adt-run as normal user, but need to run the
    container itself as root (if you use LXC system-level containers).
    This drops the hardcoded sudo commands and thus the dependency on sudo,
    which is more palatable for Debian. (LP: #1273977)
  * adt-virt-lxc: Add -e shortcut for --ephemeral, as this is an option which
    you pretty much always want.
  * adt-virt-lxc.1: Move more important options to the top.
  * adt-virt-lxc: Stop hardcoding the "ubuntu" suggested normal user; instead,
    call "getent passwd" to detect the first user with uid >= 500. Don't
    advertise suggested-normal-user if none exists.
  * adt-run: Rearrange finalizing options and reading capabilities to happen
    after testbed opening. That allows the testbed to determine its full
    capablities in its open hook.
  * adt-run: Call dpkg-source --before-build for --unbuilt-tree arguments, to
    ensure that patches are applied for "3.0 (quilt)" source trees. This is
    useful to test *-buildpackage source trees where patches don't get
    applied. (LP: #1212208)

 -- Martin Pitt <email address hidden> Fri, 31 Jan 2014 11:14:32 +0100