4 .. _device management: ../rados/operations/devices
5 .. _libstoragemgmt: https://github.com/libstorage/libstoragemgmt
10 ``ceph-volume`` scans each host in the cluster from time to time in order
11 to determine which devices are present and whether they are eligible to be
14 To print a list of devices discovered by ``cephadm``, run this command:
18 ceph orch device ls [--hostname=...] [--wide] [--refresh]
23 Hostname Path Type Serial Size Health Ident Fault Available
24 srv-01 /dev/sdb hdd 15P0A0YFFRD6 300G Unknown N/A N/A No
25 srv-01 /dev/sdc hdd 15R0A08WFRD6 300G Unknown N/A N/A No
26 srv-01 /dev/sdd hdd 15R0A07DFRD6 300G Unknown N/A N/A No
27 srv-01 /dev/sde hdd 15P0A0QDFRD6 300G Unknown N/A N/A No
28 srv-02 /dev/sdb hdd 15R0A033FRD6 300G Unknown N/A N/A No
29 srv-02 /dev/sdc hdd 15R0A05XFRD6 300G Unknown N/A N/A No
30 srv-02 /dev/sde hdd 15R0A0ANFRD6 300G Unknown N/A N/A No
31 srv-02 /dev/sdf hdd 15R0A06EFRD6 300G Unknown N/A N/A No
32 srv-03 /dev/sdb hdd 15R0A0OGFRD6 300G Unknown N/A N/A No
33 srv-03 /dev/sdc hdd 15R0A0P7FRD6 300G Unknown N/A N/A No
34 srv-03 /dev/sdd hdd 15R0A0O7FRD6 300G Unknown N/A N/A No
36 Using the ``--wide`` option provides all details relating to the device,
37 including any reasons that the device might not be eligible for use as an OSD.
39 In the above example you can see fields named "Health", "Ident", and "Fault".
40 This information is provided by integration with `libstoragemgmt`_. By default,
41 this integration is disabled (because `libstoragemgmt`_ may not be 100%
42 compatible with your hardware). To make ``cephadm`` include these fields,
43 enable cephadm's "enhanced device scan" option as follows;
47 ceph config set mgr mgr/cephadm/device_enhanced_scan true
50 Although the libstoragemgmt library performs standard SCSI inquiry calls,
51 there is no guarantee that your firmware fully implements these standards.
52 This can lead to erratic behaviour and even bus resets on some older
53 hardware. It is therefore recommended that, before enabling this feature,
54 you test your hardware's compatibility with libstoragemgmt first to avoid
55 unplanned interruptions to services.
57 There are a number of ways to test compatibility, but the simplest may be
58 to use the cephadm shell to call libstoragemgmt directly - ``cephadm shell
59 lsmcli ldl``. If your hardware is supported you should see something like
64 Path | SCSI VPD 0x83 | Link Type | Serial Number | Health Status
65 ----------------------------------------------------------------------------
66 /dev/sda | 50000396082ba631 | SAS | 15P0A0R0FRD6 | Good
67 /dev/sdb | 50000396082bbbf9 | SAS | 15P0A0YFFRD6 | Good
70 After you have enabled libstoragemgmt support, the output will look something
76 Hostname Path Type Serial Size Health Ident Fault Available
77 srv-01 /dev/sdb hdd 15P0A0YFFRD6 300G Good Off Off No
78 srv-01 /dev/sdc hdd 15R0A08WFRD6 300G Good Off Off No
81 In this example, libstoragemgmt has confirmed the health of the drives and the ability to
82 interact with the Identification and Fault LEDs on the drive enclosures. For further
83 information about interacting with these LEDs, refer to `device management`_.
86 The current release of `libstoragemgmt`_ (1.8.8) supports SCSI, SAS, and SATA based
87 local disks only. There is no official support for NVMe devices (PCIe)
89 .. _cephadm-deploy-osds:
94 Listing Storage Devices
95 -----------------------
97 In order to deploy an OSD, there must be a storage device that is *available* on
98 which the OSD will be deployed.
100 Run this command to display an inventory of storage devices on all cluster hosts:
106 A storage device is considered *available* if all of the following
109 * The device must have no partitions.
110 * The device must not have any LVM state.
111 * The device must not be mounted.
112 * The device must not contain a file system.
113 * The device must not contain a Ceph BlueStore OSD.
114 * The device must be larger than 5 GB.
116 Ceph will not provision an OSD on a device that is not available.
121 There are a few ways to create new OSDs:
123 * Tell Ceph to consume any available and unused storage device:
127 ceph orch apply osd --all-available-devices
129 * Create an OSD from a specific device on a specific host:
133 ceph orch daemon add osd *<host>*:*<device-path>*
139 ceph orch daemon add osd host1:/dev/sdb
141 Advanced OSD creation from specific devices on a specific host:
145 ceph orch daemon add osd host1:data_devices=/dev/sda,/dev/sdb,db_devices=/dev/sdc,osds_per_device=2
147 * You can use :ref:`drivegroups` to categorize device(s) based on their
148 properties. This might be useful in forming a clearer picture of which
149 devices are available to consume. Properties include device type (SSD or
150 HDD), device model names, size, and the hosts on which the devices exist:
154 ceph orch apply -i spec.yml
159 The ``--dry-run`` flag causes the orchestrator to present a preview of what
160 will happen without actually creating the OSDs.
166 ceph orch apply osd --all-available-devices --dry-run
170 NAME HOST DATA DB WAL
171 all-available-devices node1 /dev/vdb - -
172 all-available-devices node2 /dev/vdc - -
173 all-available-devices node3 /dev/vdd - -
175 .. _cephadm-osd-declarative:
180 The effect of ``ceph orch apply`` is persistent. This means that drives that
181 are added to the system after the ``ceph orch apply`` command completes will be
182 automatically found and added to the cluster. It also means that drives that
183 become available (by zapping, for example) after the ``ceph orch apply``
184 command completes will be automatically found and added to the cluster.
186 We will examine the effects of the following command:
190 ceph orch apply osd --all-available-devices
192 After running the above command:
194 * If you add new disks to the cluster, they will automatically be used to
196 * If you remove an OSD and clean the LVM physical volume, a new OSD will be
197 created automatically.
199 To disable the automatic creation of OSD on available devices, use the
200 ``unmanaged`` parameter:
202 If you want to avoid this behavior (disable automatic creation of OSD on available devices), use the ``unmanaged`` parameter:
206 ceph orch apply osd --all-available-devices --unmanaged=true
210 Keep these three facts in mind:
212 - The default behavior of ``ceph orch apply`` causes cephadm constantly to reconcile. This means that cephadm creates OSDs as soon as new drives are detected.
214 - Setting ``unmanaged: True`` disables the creation of OSDs. If ``unmanaged: True`` is set, nothing will happen even if you apply a new OSD service.
216 - ``ceph orch daemon add`` creates OSDs, but does not add an OSD service.
218 * For cephadm, see also :ref:`cephadm-spec-unmanaged`.
220 .. _cephadm-osd-removal:
225 Removing an OSD from a cluster involves two steps:
227 #. evacuating all placement groups (PGs) from the cluster
228 #. removing the PG-free OSD from the cluster
230 The following command performs these two steps:
234 ceph orch osd rm <osd_id(s)> [--replace] [--force]
244 Scheduled OSD(s) for removal
246 OSDs that are not safe to destroy will be rejected.
249 After removing OSDs, if the drives the OSDs were deployed on once again
250 become available, cephadm may automatically try to deploy more OSDs
251 on these drives if they match an existing drivegroup spec. If you deployed
252 the OSDs you are removing with a spec and don't want any new OSDs deployed on
253 the drives after removal, it's best to modify the drivegroup spec before removal.
254 Either set ``unmanaged: true`` to stop it from picking up new drives at all,
255 or modify it in some way that it no longer matches the drives used for the
256 OSDs you wish to remove. Then re-apply the spec. For more info on drivegroup
257 specs see :ref:`drivegroups`. For more info on the declarative nature of
258 cephadm in reference to deploying OSDs, see :ref:`cephadm-osd-declarative`
263 You can query the state of OSD operation with the following command:
267 ceph orch osd rm status
271 OSD_ID HOST STATE PG_COUNT REPLACE FORCE STARTED_AT
272 2 cephadm-dev done, waiting for purge 0 True False 2020-07-17 13:01:43.147684
273 3 cephadm-dev draining 17 False True 2020-07-17 13:01:45.162158
274 4 cephadm-dev started 42 False True 2020-07-17 13:01:45.162158
277 When no PGs are left on the OSD, it will be decommissioned and removed from the cluster.
280 After removing an OSD, if you wipe the LVM physical volume in the device used by the removed OSD, a new OSD will be created.
281 For more information on this, read about the ``unmanaged`` parameter in :ref:`cephadm-osd-declarative`.
286 It is possible to stop queued OSD removals by using the following command:
290 ceph orch osd rm stop <osd_id(s)>
296 ceph orch osd rm stop 4
300 Stopped OSD(s) removal
302 This resets the initial state of the OSD and takes it off the removal queue.
310 orch osd rm <osd_id(s)> --replace [--force]
316 ceph orch osd rm 4 --replace
320 Scheduled OSD(s) for replacement
322 This follows the same procedure as the procedure in the "Remove OSD" section, with
323 one exception: the OSD is not permanently removed from the CRUSH hierarchy, but is
324 instead assigned a 'destroyed' flag.
327 The new OSD that will replace the removed OSD must be created on the same host
328 as the OSD that was removed.
330 **Preserving the OSD ID**
332 The 'destroyed' flag is used to determine which OSD ids will be reused in the
335 If you use OSDSpecs for OSD deployment, your newly added disks will be assigned
336 the OSD ids of their replaced counterparts. This assumes that the new disks
337 still match the OSDSpecs.
339 Use the ``--dry-run`` flag to make certain that the ``ceph orch apply osd``
340 command does what you want it to. The ``--dry-run`` flag shows you what the
341 outcome of the command will be without making the changes you specify. When
342 you are satisfied that the command will do what you want, run the command
343 without the ``--dry-run`` flag.
347 The name of your OSDSpec can be retrieved with the command ``ceph orch ls``
349 Alternatively, you can use your OSDSpec file:
353 ceph orch apply -i <osd_spec_file> --dry-run
357 NAME HOST DATA DB WAL
358 <name_of_osd_spec> node1 /dev/vdb - -
361 When this output reflects your intention, omit the ``--dry-run`` flag to
362 execute the deployment.
365 Erasing Devices (Zapping Devices)
366 ---------------------------------
368 Erase (zap) a device so that it can be reused. ``zap`` calls ``ceph-volume
369 zap`` on the remote host.
373 ceph orch device zap <hostname> <path>
379 ceph orch device zap my_hostname /dev/sdx
382 If the unmanaged flag is unset, cephadm automatically deploys drives that
383 match the OSDSpec. For example, if you use the
384 ``all-available-devices`` option when creating OSDs, when you ``zap`` a
385 device the cephadm orchestrator automatically creates a new OSD in the
386 device. To disable this behavior, see :ref:`cephadm-osd-declarative`.
391 Automatically tuning OSD memory
392 ===============================
394 OSD daemons will adjust their memory consumption based on the
395 ``osd_memory_target`` config option (several gigabytes, by
396 default). If Ceph is deployed on dedicated nodes that are not sharing
397 memory with other services, cephadm can automatically adjust the per-OSD
398 memory consumption based on the total amount of RAM and the number of deployed
401 .. warning:: Cephadm sets ``osd_memory_target_autotune`` to ``true`` by default which is unsuitable for hyperconverged infrastructures.
403 Cephadm will start with a fraction
404 (``mgr/cephadm/autotune_memory_target_ratio``, which defaults to
405 ``.7``) of the total RAM in the system, subtract off any memory
406 consumed by non-autotuned daemons (non-OSDs, for OSDs for which
407 ``osd_memory_target_autotune`` is false), and then divide by the
410 The final targets are reflected in the config database with options like::
412 WHO MASK LEVEL OPTION VALUE
413 osd host:foo basic osd_memory_target 126092301926
414 osd host:bar basic osd_memory_target 6442450944
416 Both the limits and the current memory consumed by each daemon are visible from
417 the ``ceph orch ps`` output in the ``MEM LIMIT`` column::
419 NAME HOST PORTS STATUS REFRESHED AGE MEM USED MEM LIMIT VERSION IMAGE ID CONTAINER ID
420 osd.1 dael running (3h) 10s ago 3h 72857k 117.4G 17.0.0-3781-gafaed750 7015fda3cd67 9e183363d39c
421 osd.2 dael running (81m) 10s ago 81m 63989k 117.4G 17.0.0-3781-gafaed750 7015fda3cd67 1f0cc479b051
422 osd.3 dael running (62m) 10s ago 62m 64071k 117.4G 17.0.0-3781-gafaed750 7015fda3cd67 ac5537492f27
424 To exclude an OSD from memory autotuning, disable the autotune option
425 for that OSD and also set a specific memory target. For example,
429 ceph config set osd.123 osd_memory_target_autotune false
430 ceph config set osd.123 osd_memory_target 16G
435 Advanced OSD Service Specifications
436 ===================================
438 :ref:`orchestrator-cli-service-spec`\s of type ``osd`` are a way to describe a
439 cluster layout, using the properties of disks. Service specifications give the
440 user an abstract way to tell Ceph which disks should turn into OSDs with which
441 configurations, without knowing the specifics of device names and paths.
443 Service specifications make it possible to define a yaml or json file that can
444 be used to reduce the amount of manual work involved in creating OSDs.
446 For example, instead of running the following command:
448 .. prompt:: bash [monitor.1]#
450 ceph orch daemon add osd *<host>*:*<path-to-device>*
452 for each device and each host, we can define a yaml or json file that allows us
453 to describe the layout. Here's the most basic example.
455 Create a file called (for example) ``osd_spec.yml``:
460 service_id: default_drive_group # custom name of the osd spec
462 host_pattern: '*' # which hosts to target
464 data_devices: # the type of devices you are applying specs to
465 all: true # a filter, check below for a full list
469 #. Turn any available device (ceph-volume decides what 'available' is) into an
470 OSD on all hosts that match the glob pattern '*'. (The glob pattern matches
471 against the registered hosts from `host ls`) A more detailed section on
472 host_pattern is available below.
474 #. Then pass it to `osd create` like this:
476 .. prompt:: bash [monitor.1]#
478 ceph orch apply -i /path/to/osd_spec.yml
480 This instruction will be issued to all the matching hosts, and will deploy
483 Setups more complex than the one specified by the ``all`` filter are
484 possible. See :ref:`osd_filters` for details.
486 A ``--dry-run`` flag can be passed to the ``apply osd`` command to display a
487 synopsis of the proposed layout.
491 .. prompt:: bash [monitor.1]#
493 ceph orch apply -i /path/to/osd_spec.yml --dry-run
503 Filters are applied using an `AND` gate by default. This means that a drive
504 must fulfill all filter criteria in order to get selected. This behavior can
505 be adjusted by setting ``filter_logic: OR`` in the OSD specification.
507 Filters are used to assign disks to groups, using their attributes to group
510 The attributes are based off of ceph-volume's disk query. You can retrieve
511 information about the attributes with this command:
515 ceph-volume inventory </path/to/disk>
520 Specific disks can be targeted by vendor or model:
524 model: disk_model_name
530 vendor: disk_vendor_name
536 Specific disks can be targeted by `Size`:
545 Size specifications can be of the following forms:
554 To include disks of an exact size
560 To include disks within a given range of size:
566 To include disks that are less than or equal to 10G in size:
572 To include disks equal to or greater than 40G in size:
578 Sizes don't have to be specified exclusively in Gigabytes(G).
580 Other units of size are supported: Megabyte(M), Gigabyte(G) and Terrabyte(T).
581 Appending the (B) for byte is also supported: ``MB``, ``GB``, ``TB``.
587 This operates on the 'rotational' attribute of the disk.
593 `1` to match all disks that are rotational
595 `0` to match all disks that are non-rotational (SSD, NVME etc)
601 This will take all disks that are 'available'
603 .. note:: This is exclusive for the data_devices section.
613 If you have specified some valid filters but want to limit the number of disks that they match, use the ``limit`` directive:
619 For example, if you used `vendor` to match all disks that are from `VendorA`
620 but want to use only the first two, you could use `limit`:
628 .. note:: `limit` is a last resort and shouldn't be used if it can be avoided.
634 There are multiple optional settings you can use to change the way OSDs are deployed.
635 You can add these options to the base level of an OSD spec for it to take effect.
637 This example would deploy all OSDs with encryption enabled.
642 service_id: example_osd_spec
650 See a full list in the DriveGroupSpecs
652 .. py:currentmodule:: ceph.deployment.drive_group
654 .. autoclass:: DriveGroupSpec
656 :exclude-members: from_json
664 All nodes with the same setup
678 This is a common setup and can be described quite easily:
683 service_id: osd_spec_default
688 model: HDD-123-foo # Note, HDD-123 would also be valid
690 model: MC-55-44-XZ # Same here, MC-55-44 is valid
692 However, we can improve it by reducing the filters on core properties of the drives:
697 service_id: osd_spec_default
706 Now, we enforce all rotating devices to be declared as 'data devices' and all non-rotating devices will be used as shared_devices (wal, db)
708 If you know that drives with more than 2TB will always be the slower data devices, you can also filter by size:
713 service_id: osd_spec_default
722 .. note:: All of the above OSD specs are equally valid. Which of those you want to use depends on taste and on how much you expect your node layout to change.
725 Multiple OSD specs for a single host
726 ------------------------------------
728 Here we have two distinct setups
748 * 20 HDDs should share 2 SSDs
749 * 10 SSDs should share 2 NVMes
751 This can be described with two layouts.
756 service_id: osd_spec_hdd
764 limit: 2 # db_slots is actually to be favoured here, but it's not implemented yet
767 service_id: osd_spec_ssd
776 This would create the desired layout by using all HDDs as data_devices with two SSD assigned as dedicated db/wal devices.
777 The remaining SSDs(8) will be data_devices that have the 'VendorC' NVMEs assigned as dedicated db/wal devices.
779 Multiple hosts with the same disk layout
780 ----------------------------------------
782 Assuming the cluster has different kinds of hosts each with similar disk
783 layout, it is recommended to apply different OSD specs matching only one
784 set of hosts. Typically you will have a spec for multiple hosts with the
787 The service id as the unique key: In case a new OSD spec with an already
788 applied service id is applied, the existing OSD spec will be superseded.
789 cephadm will now create new OSD daemons based on the new spec
790 definition. Existing OSD daemons will not be affected. See :ref:`cephadm-osd-declarative`.
818 You can use the 'placement' key in the layout to target certain nodes.
823 service_id: disk_layout_a
833 service_id: disk_layout_b
842 This applies different OSD specs to different hosts depending on the `placement` key.
843 See :ref:`orchestrator-cli-placement-spec`
847 Assuming each host has a unique disk layout, each OSD
848 spec needs to have a different service id
854 All previous cases co-located the WALs with the DBs.
855 It's however possible to deploy the WAL on a dedicated device as well, if it makes sense.
875 The OSD spec for this case would look like the following (using the `model` filter):
880 service_id: osd_spec_default
892 It is also possible to specify directly device paths in specific hosts like the following:
897 service_id: osd_using_paths
914 This can easily be done with other filters, like `size` or `vendor` as well.
916 .. _cephadm-osd-activate:
918 Activate existing OSDs
919 ======================
921 In case the OS of a host was reinstalled, existing OSDs need to be activated
922 again. For this use case, cephadm provides a wrapper for :ref:`ceph-volume-lvm-activate` that
923 activates all existing OSDs on a host.
927 ceph cephadm osd activate <host>...
929 This will scan all existing disks for OSDs and deploy corresponding daemons.