update to 4.2-rc1

[Michael Hudson-Doyle]

We should update to the current upstream rc. It should be fairly simple as most of the patches in the package are backports from upstream anyway.

[Launchpad Janitor (janitor)]

This bug was fixed in the package mdadm - 4.2~rc1-1ubuntu1

---------------
mdadm (4.2~rc1-1ubuntu1) impish; urgency=medium

  * Merge from Debian experimental (LP: #1935895), remaining changes:
    - Ship finalrd hook.
    - Do not install mdadm-shutdown.service on Ubuntu.
    - Drop broken and unused init scripts in favor of native systemd units,
      which can cause failure to reconfigure mdadm package under certain
      confiment types.
    - Drop /etc/cron.d/mdadm and migrate to systemd mdcheck_start|continue
      timer units.
    - Drop /etc/cron.daily/mdadm and migrate to system mdmonitor-oneshot
      timer unit.
    - mdcheck_start.timer configures the mdcheck on a first sunday of the
      month, with a randomized start delay of up to 24h, and runs for at
      most 6h. mdcheck_continue.timer kicks off daily, with a randomized
      start delay of up to 12h, and continues mdcheck for at most 6h.
    - mdmonitor-oneshot.timer runs daily, with a randomized start delay of
      up to 24h.
    - One can use systemd drop-ins to change .timer units timings, set
      environmental variables to decrease/increase the length of checking,
      or start the checks by hand. Previously used checkarray is still
      available, albeit not used by timer units.
    - Above ensures that previous daily / monthly checks are performed, but
      are randomized, such that performance is not as impacted across a
      cluster of machines.
    - Honor the debconf daily autoscan setting in the systemd timer.
    - d/p/mdcheck-fix-empty-spaces-in-timer-unit-files.patch:
      - The spaces were causing a FTBFS situation, triggered by
        dh_installsystemd not finding units.
    - Mention mdmonitor-oneshot.service instead of /etc/cron.daily/mdadm
      in /etc/default/mdadm
  * Dropped changes:
    - d/p/mdcheck-when-mdcheck-start-is-enabled-mdcheck-continue-too.patch:
      - When mdcheck_start is enabled, enable mdcheck_continue too.
    - d/p/mdcheck-log-when-done.patch:
      - Make sure mdcheck logs the completion too, so that
      it can be determined how long the raid check took.
    - Cherrypick upstream fix to allow assembling dirty degrated IMSM raid
      arrays with --force flag

mdadm (4.2~rc1-1) experimental; urgency=medium

  * New release candidate from upstream.
  * Add libudev-dev to Build-Depends.
  * Drop patches previously cherry-picked from upstream:
    - 0001-Assemble-keep-MD_DISK_FAILFAST-and-MD_DISK_WRITEMOST.patch
      (now 0833f9c3)
    - 0002-Document-PART-POLICY-lines.patch (now 6b611284)
    - 0003-policy-support-devices-with-multiple-paths.patch (now cd72f9d1)
    - 0004-mdcheck-add-systemd-unit-files-to-run-mdcheck.patch (now 4199d3c6)
    - 0005-Monitor-add-system-timer-to-run-oneshot-periodically.patch
      (now 7cd7e91a)
    - 0006-imsm-update-metadata-correctly-while-raid10-double-d.patch
      (now d7a1fda2)
    - 0007-Assemble-mask-FAILFAST-and-WRITEMOSTLY-flags-when-fi.patch
      (now 563ac108)
    - 0008-Grow-avoid-overflow-in-compute_backup_blocks.patch
      (now 085df422)
    - 0009-Grow-report-correct-new-chunk-size.patch (now 76d505de)
    - 0010-policy.c-prevent-NULL-pointer-referencing.patch (now 467...