ceph/doc/ceph-volume/lvm/activate.rst

   1 .. _ceph-volume-lvm-activate:
   2
   3 ``activate``
   4 ============
   5 Once :ref:`ceph-volume-lvm-prepare` is completed, and all the various steps
   6 that entails are done, the volume is ready to get "activated".
   7
   8 This activation process enables a systemd unit that persists the OSD ID and its
   9 UUID (also called ``fsid`` in Ceph CLI tools), so that at boot time it can
  10 understand what OSD is enabled and needs to be mounted.
  11
  12 .. note:: The execution of this call is fully idempotent, and there is no
  13           side-effects when running multiple times
  14
  15 New OSDs
  16 --------
  17 To activate newly prepared OSDs both the :term:`OSD id` and :term:`OSD uuid`
  18 need to be supplied. For example::
  19
  20     ceph-volume lvm activate --bluestore 0 0263644D-0BF1-4D6D-BC34-28BD98AE3BC8
  21
  22 .. note:: The UUID is stored in the ``fsid`` file in the OSD path, which is
  23           generated when :ref:`ceph-volume-lvm-prepare` is used.
  24
  25 Activating all OSDs
  26 -------------------
  27 It is possible to activate all existing OSDs at once by using the ``--all``
  28 flag. For example::
  29
  30     ceph-volume lvm activate --all
  31
  32 This call will inspect all the OSDs created by ceph-volume that are inactive
  33 and will activate them one by one. If any of the OSDs are already running, it
  34 will report them in the command output and skip them, making it safe to rerun
  35 (idempotent).
  36
  37 requiring uuids
  38 ^^^^^^^^^^^^^^^
  39 The :term:`OSD uuid` is being required as an extra step to ensure that the
  40 right OSD is being activated. It is entirely possible that a previous OSD with
  41 the same id exists and would end up activating the incorrect one.
  42
  43
  44 dmcrypt
  45 ^^^^^^^
  46 If the OSD was prepared with dmcrypt by ceph-volume, there is no need to
  47 specify ``--dmcrypt`` on the command line again (that flag is not available for
  48 the ``activate`` subcommand). An encrypted OSD will be automatically detected.
  49
  50
  51 Discovery
  52 ---------
  53 With OSDs previously created by ``ceph-volume``, a *discovery* process is
  54 performed using :term:`LVM tags` to enable the systemd units.
  55
  56 The systemd unit will capture the :term:`OSD id` and :term:`OSD uuid` and
  57 persist it. Internally, the activation will enable it like::
  58
  59     systemctl enable ceph-volume@lvm-$id-$uuid
  60
  61 For example::
  62
  63     systemctl enable ceph-volume@lvm-0-8715BEB4-15C5-49DE-BA6F-401086EC7B41
  64
  65 Would start the discovery process for the OSD with an id of ``0`` and a UUID of
  66 ``8715BEB4-15C5-49DE-BA6F-401086EC7B41``.
  67
  68 .. note:: for more details on the systemd workflow see :ref:`ceph-volume-lvm-systemd`
  69
  70 The systemd unit will look for the matching OSD device, and by looking at its
  71 :term:`LVM tags` will proceed to:
  72
  73 # mount the device in the corresponding location (by convention this is
  74   ``/var/lib/ceph/osd/<cluster name>-<osd id>/``)
  75
  76 # ensure that all required devices are ready for that OSD. In the case of
  77 a journal (when ``--filestore`` is selected) the device will be queried (with
  78 ``blkid`` for partitions, and lvm for logical volumes) to ensure that the
  79 correct device is being linked. The symbolic link will *always* be re-done to
  80 ensure that the correct device is linked.
  81
  82 # start the ``ceph-osd@0`` systemd unit
  83
  84 .. note:: The system infers the objectstore type (filestore or bluestore) by
  85           inspecting the LVM tags applied to the OSD devices
  86
  87 Existing OSDs
  88 -------------
  89 For existing OSDs that have been deployed with ``ceph-disk``, they need to be
  90 scanned and activated :ref:`using the simple sub-command <ceph-volume-simple>`.
  91 If a different tooling was used then the only way to port them over to the new
  92 mechanism is to prepare them again (losing data). See
  93 :ref:`ceph-volume-lvm-existing-osds` for details on how to proceed.
  94
  95 Summary
  96 -------
  97 To recap the ``activate`` process for :term:`bluestore`:
  98
  99 #. require both :term:`OSD id` and :term:`OSD uuid`
 100 #. enable the system unit with matching id and uuid
 101 #. Create the ``tmpfs`` mount at the OSD directory in
 102    ``/var/lib/ceph/osd/$cluster-$id/``
 103 #. Recreate all the files needed with ``ceph-bluestore-tool prime-osd-dir`` by
 104    pointing it to the OSD ``block`` device.
 105 #. the systemd unit will ensure all devices are ready and linked
 106 #. the matching ``ceph-osd`` systemd unit will get started
 107
 108 And for :term:`filestore`:
 109
 110 #. require both :term:`OSD id` and :term:`OSD uuid`
 111 #. enable the system unit with matching id and uuid
 112 #. the systemd unit will ensure all devices are ready and mounted (if needed)
 113 #. the matching ``ceph-osd`` systemd unit will get started