1CEPH(8) Ceph CEPH(8)
2
6 ceph - ceph administration tool
7
9 ceph auth [ add | caps | del | export | get | get-key | get-or-create | get-or-create-key | import | list | print-key | print_key ] ...
10 ceph compact
12 ceph config [ dump | ls | help | get | show | show-with-defaults | set | rm | log | reset | assimilate-conf | generate-minimal-conf ] ...
14 ceph config-key [ rm | exists | get | ls | dump | set ] ...
16 ceph daemon <name> | <path> <command> ...
18 ceph daemonperf <name> | <path> [ interval [ count ] ]
20 ceph df {detail}
22 ceph fs [ ls | new | reset | rm | authorize ] ...
24 ceph fsid
26 ceph health {detail}
28 ceph injectargs <injectedargs> [ <injectedargs>... ]
30 ceph log <logtext> [ <logtext>... ]
32 ceph mds [ compat | fail | rm | rmfailed | set_state | stat | repaired ] ...
34 ceph mon [ add | dump | getmap | remove | stat ] ...
36 ceph osd [ blocklist | blocked-by | create | new | deep-scrub | df | down | dump | erasure-code-profile | find | getcrushmap | getmap | getmaxosd | in | ls | lspools | map | metadata | ok-to-stop | out | pause | perf | pg-temp | force-create-pg | primary-affinity | primary-temp | repair | reweight | reweight-by-pg | rm | destroy | purge | safe-to-destroy | scrub | set | setcrushmap | setmaxosd | stat | tree | unpause | unset ] ...
38 ceph osd crush [ add | add-bucket | create-or-move | dump | get-tunable | link | move | remove | rename-bucket | reweight | reweight-all | reweight-subtree | rm | rule | set | set-tunable | show-tunables | tunables | unlink ] ...
40 ceph osd pool [ create | delete | get | get-quota | ls | mksnap | rename | rmsnap | set | set-quota | stats ] ...
42 ceph osd pool application [ disable | enable | get | rm | set ] ...
44 ceph osd tier [ add | add-cache | cache-mode | remove | remove-overlay | set-overlay ] ...
46 ceph pg [ debug | deep-scrub | dump | dump_json | dump_pools_json | dump_stuck | getmap | ls | ls-by-osd | ls-by-pool | ls-by-primary | map | repair | scrub | stat ] ...
48 ceph quorum_status
50 ceph report { <tags> [ <tags>... ] }
52 ceph status
54 ceph sync force {--yes-i-really-mean-it} {--i-know-what-i-am-doing}
56 ceph tell <name (type.id)> <command> [options...]
58 ceph version
60
63 ceph is a control utility which is used for manual deployment and main‐
64 tenance of a Ceph cluster. It provides a diverse set of commands that
65 allows deployment of monitors, OSDs, placement groups, MDS and overall
66 maintenance, administration of the cluster.
67
69 auth
70 Manage authentication keys. It is used for adding, removing, exporting
71 or updating of authentication keys for a particular entity such as a
72 monitor or OSD. It uses some additional subcommands.
73 Subcommand add adds authentication info for a particular entity from
75 input file, or random key if no input is given and/or any caps speci‐
76 fied in the command.
77 Usage:
79 ceph auth add <entity> {<caps> [<caps>...]}
81 Subcommand caps updates caps for name from caps specified in the com‐
83 mand.
84 Usage:
86 ceph auth caps <entity> <caps> [<caps>...]
88 Subcommand del deletes all caps for name.
90 Usage:
92 ceph auth del <entity>
94 Subcommand export writes keyring for requested entity, or master
96 keyring if none given.
97 Usage:
99 ceph auth export {<entity>}
101 Subcommand get writes keyring file with requested key.
103 Usage:
105 ceph auth get <entity>
107 Subcommand get-key displays requested key.
109 Usage:
111 ceph auth get-key <entity>
113 Subcommand get-or-create adds authentication info for a particular en‐
115 tity from input file, or random key if no input given and/or any caps
116 specified in the command.
117 Usage:
119 ceph auth get-or-create <entity> {<caps> [<caps>...]}
121 Subcommand get-or-create-key gets or adds key for name from system/caps
123 pairs specified in the command. If key already exists, any given caps
124 must match the existing caps for that key.
125 Usage:
127 ceph auth get-or-create-key <entity> {<caps> [<caps>...]}
129 Subcommand import reads keyring from input file.
131 Usage:
133 ceph auth import
135 Subcommand ls lists authentication state.
137 Usage:
139 ceph auth ls
141 Subcommand print-key displays requested key.
143 Usage:
145 ceph auth print-key <entity>
147 Subcommand print_key displays requested key.
149 Usage:
151 ceph auth print_key <entity>
153 compact
155 Causes compaction of monitor's leveldb storage.
156 Usage:
158 ceph compact
160 config
162 Configure the cluster. By default, Ceph daemons and clients retrieve
163 their configuration options from monitor when they start, and are up‐
164 dated if any of the tracked options is changed at run time. It uses
165 following additional subcommand.
166 Subcommand dump to dump all options for the cluster
168 Usage:
170 ceph config dump
172 Subcommand ls to list all option names for the cluster
174 Usage:
176 ceph config ls
178 Subcommand help to describe the specified configuration option
180 Usage:
182 ceph config help <option>
184 Subcommand get to dump the option(s) for the specified entity.
186 Usage:
188 ceph config get <who> {<option>}
190 Subcommand show to display the running configuration of the specified
192 entity. Please note, unlike get, which only shows the options managed
193 by monitor, show displays all the configurations being actively used.
194 These options are pulled from several sources, for instance, the com‐
195 piled-in default value, the monitor's configuration database, ceph.conf
196 file on the host. The options can even be overridden at runtime. So,
197 there is chance that the configuration options in the output of show
198 could be different from those in the output of get.
199 Usage:
201 ceph config show {<who>}
203 Subcommand show-with-defaults to display the running configuration
205 along with the compiled-in defaults of the specified entity
206 Usage:
208 ceph config show {<who>}
210 Subcommand set to set an option for one or more specified entities
212 Usage:
214 ceph config set <who> <option> <value> {--force}
216 Subcommand rm to clear an option for one or more entities
218 Usage:
220 ceph config rm <who> <option>
222 Subcommand log to show recent history of config changes. If count op‐
224 tion is omitted it defeaults to 10.
225 Usage:
227 ceph config log {<count>}
229 Subcommand reset to revert configuration to the specified historical
231 version
232 Usage:
234 ceph config reset <version>
236 Subcommand assimilate-conf to assimilate options from stdin, and return
238 a new, minimal conf file
239 Usage:
241 ceph config assimilate-conf -i <input-config-path> > <output-config-path>
243 ceph config assimilate-conf < <input-config-path>
244 Subcommand generate-minimal-conf to generate a minimal ceph.conf file,
246 which can be used for bootstrapping a daemon or a client.
247 Usage:
249 ceph config generate-minimal-conf > <minimal-config-path>
251 config-key
253 Manage configuration key. Config-key is a general purpose key/value
254 service offered by the monitors. This service is mainly used by Ceph
255 tools and daemons for persisting various settings. Among which,
256 ceph-mgr modules uses it for storing their options. It uses some addi‐
257 tional subcommands.
258 Subcommand rm deletes configuration key.
260 Usage:
262 ceph config-key rm <key>
264 Subcommand exists checks for configuration keys existence.
266 Usage:
268 ceph config-key exists <key>
270 Subcommand get gets the configuration key.
272 Usage:
274 ceph config-key get <key>
276 Subcommand ls lists configuration keys.
278 Usage:
280 ceph config-key ls
282 Subcommand dump dumps configuration keys and values.
284 Usage:
286 ceph config-key dump
288 Subcommand set puts configuration key and value.
290 Usage:
292 ceph config-key set <key> {<val>}
294 daemon
296 Submit admin-socket commands.
297 Usage:
299 ceph daemon {daemon_name|socket_path} {command} ...
301 Example:
303 ceph daemon osd.0 help
305 daemonperf
307 Watch performance counters from a Ceph daemon.
308 Usage:
310 ceph daemonperf {daemon_name|socket_path} [{interval} [{count}]]
312 df
314 Show cluster's free space status.
315 Usage:
317 ceph df {detail}
319 features
321 Show the releases and features of all connected daemons and clients
322 connected to the cluster, along with the numbers of them in each bucket
323 grouped by the corresponding features/releases. Each release of Ceph
324 supports a different set of features, expressed by the features bit‐
325 mask. New cluster features require that clients support the feature, or
326 else they are not allowed to connect to these new features. As new fea‐
327 tures or capabilities are enabled after an upgrade, older clients are
328 prevented from connecting.
329 Usage:
331 ceph features
333 fs
335 Manage cephfs file systems. It uses some additional subcommands.
336 Subcommand ls to list file systems
338 Usage:
340 ceph fs ls
342 Subcommand new to make a new file system using named pools <metadata>
344 and <data>
345 Usage:
347 ceph fs new <fs_name> <metadata> <data>
349 Subcommand reset is used for disaster recovery only: reset to a sin‐
351 gle-MDS map
352 Usage:
354 ceph fs reset <fs_name> {--yes-i-really-mean-it}
356 Subcommand rm to disable the named file system
358 Usage:
360 ceph fs rm <fs_name> {--yes-i-really-mean-it}
362 Subcommand authorize creates a new client that will be authorized for
364 the given path in <fs_name>. Pass / to authorize for the entire FS.
365 <perms> below can be r, rw or rwp.
366 Usage:
368 ceph fs authorize <fs_name> client.<client_id> <path> <perms> [<path> <perms>...]
370 fsid
372 Show cluster's FSID/UUID.
373 Usage:
375 ceph fsid
377 health
379 Show cluster's health.
380 Usage:
382 ceph health {detail}
384 heap
386 Show heap usage info (available only if compiled with tcmalloc)
387 Usage:
389 ceph tell <name (type.id)> heap dump|start_profiler|stop_profiler|stats
391 Subcommand release to make TCMalloc to releases no-longer-used memory
393 back to the kernel at once.
394 Usage:
396 ceph tell <name (type.id)> heap release
398 Subcommand (get|set)_release_rate get or set the TCMalloc memory re‐
400 lease rate. TCMalloc releases no-longer-used memory back to the kernel
401 gradually. the rate controls how quickly this happens. Increase this
402 setting to make TCMalloc to return unused memory more frequently. 0
403 means never return memory to system, 1 means wait for 1000 pages after
404 releasing a page to system. It is 1.0 by default..
405 Usage:
407 ceph tell <name (type.id)> heap get_release_rate|set_release_rate {<val>}
409 injectargs
411 Inject configuration arguments into monitor.
412 Usage:
414 ceph injectargs <injected_args> [<injected_args>...]
416 log
418 Log supplied text to the monitor log.
419 Usage:
421 ceph log <logtext> [<logtext>...]
423 mds
425 Manage metadata server configuration and administration. It uses some
426 additional subcommands.
427 Subcommand compat manages compatible features. It uses some additional
429 subcommands.
430 Subcommand rm_compat removes compatible feature.
432 Usage:
434 ceph mds compat rm_compat <int[0-]>
436 Subcommand rm_incompat removes incompatible feature.
438 Usage:
440 ceph mds compat rm_incompat <int[0-]>
442 Subcommand show shows mds compatibility settings.
444 Usage:
446 ceph mds compat show
448 Subcommand fail forces mds to status fail.
450 Usage:
452 ceph mds fail <role|gid>
454 Subcommand rm removes inactive mds.
456 Usage:
458 ceph mds rm <int[0-]> <name> (type.id)>
460 Subcommand rmfailed removes failed mds.
462 Usage:
464 ceph mds rmfailed <int[0-]>
466 Subcommand set_state sets mds state of <gid> to <numeric-state>.
468 Usage:
470 ceph mds set_state <int[0-]> <int[0-20]>
472 Subcommand stat shows MDS status.
474 Usage:
476 ceph mds stat
478 Subcommand repaired mark a damaged MDS rank as no longer damaged.
480 Usage:
482 ceph mds repaired <role>
484 mon
486 Manage monitor configuration and administration. It uses some addi‐
487 tional subcommands.
488 Subcommand add adds new monitor named <name> at <addr>.
490 Usage:
492 ceph mon add <name> <IPaddr[:port]>
494 Subcommand dump dumps formatted monmap (optionally from epoch)
496 Usage:
498 ceph mon dump {<int[0-]>}
500 Subcommand getmap gets monmap.
502 Usage:
504 ceph mon getmap {<int[0-]>}
506 Subcommand remove removes monitor named <name>.
508 Usage:
510 ceph mon remove <name>
512 Subcommand stat summarizes monitor status.
514 Usage:
516 ceph mon stat
518 mgr
520 Ceph manager daemon configuration and management.
521 Subcommand dump dumps the latest MgrMap, which describes the active and
523 standby manager daemons.
524 Usage:
526 ceph mgr dump
528 Subcommand fail will mark a manager daemon as failed, removing it from
530 the manager map. If it is the active manager daemon a standby will
531 take its place.
532 Usage:
534 ceph mgr fail <name>
536 Subcommand module ls will list currently enabled manager modules (plug‐
538 ins).
539 Usage:
541 ceph mgr module ls
543 Subcommand module enable will enable a manager module. Available mod‐
545 ules are included in MgrMap and visible via mgr dump.
546 Usage:
548 ceph mgr module enable <module>
550 Subcommand module disable will disable an active manager module.
552 Usage:
554 ceph mgr module disable <module>
556 Subcommand metadata will report metadata about all manager daemons or,
558 if the name is specified, a single manager daemon.
559 Usage:
561 ceph mgr metadata [name]
563 Subcommand versions will report a count of running daemon versions.
565 Usage:
567 ceph mgr versions
569 Subcommand count-metadata will report a count of any daemon metadata
571 field.
572 Usage:
574 ceph mgr count-metadata <field>
576 osd
578 Manage OSD configuration and administration. It uses some additional
579 subcommands.
580 Subcommand blocklist manage blocklisted clients. It uses some addi‐
582 tional subcommands.
583 Subcommand add add <addr> to blocklist (optionally until <expire> sec‐
585 onds from now)
586 Usage:
588 ceph osd blocklist add <EntityAddr> {<float[0.0-]>}
590 Subcommand ls show blocklisted clients
592 Usage:
594 ceph osd blocklist ls
596 Subcommand rm remove <addr> from blocklist
598 Usage:
600 ceph osd blocklist rm <EntityAddr>
602 Subcommand blocked-by prints a histogram of which OSDs are blocking
604 their peers
605 Usage:
607 ceph osd blocked-by
609 Subcommand create creates new osd (with optional UUID and ID).
611 This command is DEPRECATED as of the Luminous release, and will be re‐
613 moved in a future release.
614 Subcommand new should instead be used.
616 Usage:
618 ceph osd create {<uuid>} {<id>}
620 Subcommand new can be used to create a new OSD or to recreate a previ‐
622 ously destroyed OSD with a specific id. The new OSD will have the spec‐
623 ified uuid, and the command expects a JSON file containing the base64
624 cephx key for auth entity client.osd.<id>, as well as optional base64
625 cepx key for dm-crypt lockbox access and a dm-crypt key. Specifying a
626 dm-crypt requires specifying the accompanying lockbox cephx key.
627 Usage:
629 ceph osd new {<uuid>} {<id>} -i {<params.json>}
631 The parameters JSON file is optional but if provided, is expected to
633 maintain a form of the following format:
634 {
636 "cephx_secret": "AQBWtwhZdBO5ExAAIDyjK2Bh16ZXylmzgYYEjg==",
637 "crush_device_class": "myclass"
638 }
639 Or:
641 {
643 "cephx_secret": "AQBWtwhZdBO5ExAAIDyjK2Bh16ZXylmzgYYEjg==",
644 "cephx_lockbox_secret": "AQDNCglZuaeVCRAAYr76PzR1Anh7A0jswkODIQ==",
645 "dmcrypt_key": "<dm-crypt key>",
646 "crush_device_class": "myclass"
647 }
648 Or:
650 {
652 "crush_device_class": "myclass"
653 }
654 The "crush_device_class" property is optional. If specified, it will
656 set the initial CRUSH device class for the new OSD.
657 Subcommand crush is used for CRUSH management. It uses some additional
659 subcommands.
660 Subcommand add adds or updates crushmap position and weight for <name>
662 with <weight> and location <args>.
663 Usage:
665 ceph osd crush add <osdname (id|osd.id)> <float[0.0-]> <args> [<args>...]
667 Subcommand add-bucket adds no-parent (probably root) crush bucket
669 <name> of type <type>.
670 Usage:
672 ceph osd crush add-bucket <name> <type>
674 Subcommand create-or-move creates entry or moves existing entry for
676 <name> <weight> at/to location <args>.
677 Usage:
679 ceph osd crush create-or-move <osdname (id|osd.id)> <float[0.0-]> <args>
681 [<args>...]
682 Subcommand dump dumps crush map.
684 Usage:
686 ceph osd crush dump
688 Subcommand get-tunable get crush tunable straw_calc_version
690 Usage:
692 ceph osd crush get-tunable straw_calc_version
694 Subcommand link links existing entry for <name> under location <args>.
696 Usage:
698 ceph osd crush link <name> <args> [<args>...]
700 Subcommand move moves existing entry for <name> to location <args>.
702 Usage:
704 ceph osd crush move <name> <args> [<args>...]
706 Subcommand remove removes <name> from crush map (everywhere, or just at
708 <ancestor>).
709 Usage:
711 ceph osd crush remove <name> {<ancestor>}
713 Subcommand rename-bucket renames bucket <srcname> to <dstname>
715 Usage:
717 ceph osd crush rename-bucket <srcname> <dstname>
719 Subcommand reweight change <name>'s weight to <weight> in crush map.
721 Usage:
723 ceph osd crush reweight <name> <float[0.0-]>
725 Subcommand reweight-all recalculate the weights for the tree to ensure
727 they sum correctly
728 Usage:
730 ceph osd crush reweight-all
732 Subcommand reweight-subtree changes all leaf items beneath <name> to
734 <weight> in crush map
735 Usage:
737 ceph osd crush reweight-subtree <name> <weight>
739 Subcommand rm removes <name> from crush map (everywhere, or just at
741 <ancestor>).
742 Usage:
744 ceph osd crush rm <name> {<ancestor>}
746 Subcommand rule is used for creating crush rules. It uses some addi‐
748 tional subcommands.
749 Subcommand create-erasure creates crush rule <name> for erasure coded
751 pool created with <profile> (default default).
752 Usage:
754 ceph osd crush rule create-erasure <name> {<profile>}
756 Subcommand create-simple creates crush rule <name> to start from
758 <root>, replicate across buckets of type <type>, using a choose mode of
759 <firstn|indep> (default firstn; indep best for erasure pools).
760 Usage:
762 ceph osd crush rule create-simple <name> <root> <type> {firstn|indep}
764 Subcommand dump dumps crush rule <name> (default all).
766 Usage:
768 ceph osd crush rule dump {<name>}
770 Subcommand ls lists crush rules.
772 Usage:
774 ceph osd crush rule ls
776 Subcommand rm removes crush rule <name>.
778 Usage:
780 ceph osd crush rule rm <name>
782 Subcommand set used alone, sets crush map from input file.
784 Usage:
786 ceph osd crush set
788 Subcommand set with osdname/osd.id update crushmap position and weight
790 for <name> to <weight> with location <args>.
791 Usage:
793 ceph osd crush set <osdname (id|osd.id)> <float[0.0-]> <args> [<args>...]
795 Subcommand set-tunable set crush tunable <tunable> to <value>. The
797 only tunable that can be set is straw_calc_version.
798 Usage:
800 ceph osd crush set-tunable straw_calc_version <value>
802 Subcommand show-tunables shows current crush tunables.
804 Usage:
806 ceph osd crush show-tunables
808 Subcommand tree shows the crush buckets and items in a tree view.
810 Usage:
812 ceph osd crush tree
814 Subcommand tunables sets crush tunables values to <profile>.
816 Usage:
818 ceph osd crush tunables legacy|argonaut|bobtail|firefly|hammer|optimal|default
820 Subcommand unlink unlinks <name> from crush map (everywhere, or just at
822 <ancestor>).
823 Usage:
825 ceph osd crush unlink <name> {<ancestor>}
827 Subcommand df shows OSD utilization
829 Usage:
831 ceph osd df {plain|tree}
833 Subcommand deep-scrub initiates deep scrub on specified osd.
835 Usage:
837 ceph osd deep-scrub <who>
839 Subcommand down sets osd(s) <id> [<id>...] down.
841 Usage:
843 ceph osd down <ids> [<ids>...]
845 Subcommand dump prints summary of OSD map.
847 Usage:
849 ceph osd dump {<int[0-]>}
851 Subcommand erasure-code-profile is used for managing the erasure code
853 profiles. It uses some additional subcommands.
854 Subcommand get gets erasure code profile <name>.
856 Usage:
858 ceph osd erasure-code-profile get <name>
860 Subcommand ls lists all erasure code profiles.
862 Usage:
864 ceph osd erasure-code-profile ls
866 Subcommand rm removes erasure code profile <name>.
868 Usage:
870 ceph osd erasure-code-profile rm <name>
872 Subcommand set creates erasure code profile <name> with [<key[=value]>
874 ...] pairs. Add a --force at the end to override an existing profile
875 (IT IS RISKY).
876 Usage:
878 ceph osd erasure-code-profile set <name> {<profile> [<profile>...]}
880 Subcommand find find osd <id> in the CRUSH map and shows its location.
882 Usage:
884 ceph osd find <int[0-]>
886 Subcommand getcrushmap gets CRUSH map.
888 Usage:
890 ceph osd getcrushmap {<int[0-]>}
892 Subcommand getmap gets OSD map.
894 Usage:
896 ceph osd getmap {<int[0-]>}
898 Subcommand getmaxosd shows largest OSD id.
900 Usage:
902 ceph osd getmaxosd
904 Subcommand in sets osd(s) <id> [<id>...] in.
906 Usage:
908 ceph osd in <ids> [<ids>...]
910 Subcommand lost marks osd as permanently lost. THIS DESTROYS DATA IF NO
912 MORE REPLICAS EXIST, BE CAREFUL.
913 Usage:
915 ceph osd lost <int[0-]> {--yes-i-really-mean-it}
917 Subcommand ls shows all OSD ids.
919 Usage:
921 ceph osd ls {<int[0-]>}
923 Subcommand lspools lists pools.
925 Usage:
927 ceph osd lspools {<int>}
929 Subcommand map finds pg for <object> in <pool>.
931 Usage:
933 ceph osd map <poolname> <objectname>
935 Subcommand metadata fetches metadata for osd <id>.
937 Usage:
939 ceph osd metadata {int[0-]} (default all)
941 Subcommand out sets osd(s) <id> [<id>...] out.
943 Usage:
945 ceph osd out <ids> [<ids>...]
947 Subcommand ok-to-stop checks whether the list of OSD(s) can be stopped
949 without immediately making data unavailable. That is, all data should
950 remain readable and writeable, although data redundancy may be reduced
951 as some PGs may end up in a degraded (but active) state. It will re‐
952 turn a success code if it is okay to stop the OSD(s), or an error code
953 and informative message if it is not or if no conclusion can be drawn
954 at the current time. When --max <num> is provided, up to <num> OSDs
955 IDs will return (including the provided OSDs) that can all be stopped
956 simultaneously. This allows larger sets of stoppable OSDs to be gener‐
957 ated easily by providing a single starting OSD and a max. Additional
958 OSDs are drawn from adjacent locations in the CRUSH hierarchy.
959 Usage:
961 ceph osd ok-to-stop <id> [<ids>...] [--max <num>]
963 Subcommand pause pauses osd.
965 Usage:
967 ceph osd pause
969 Subcommand perf prints dump of OSD perf summary stats.
971 Usage:
973 ceph osd perf
975 Subcommand pg-temp set pg_temp mapping pgid:[<id> [<id>...]] (develop‐
977 ers only).
978 Usage:
980 ceph osd pg-temp <pgid> {<id> [<id>...]}
982 Subcommand force-create-pg forces creation of pg <pgid>.
984 Usage:
986 ceph osd force-create-pg <pgid>
988 Subcommand pool is used for managing data pools. It uses some addi‐
990 tional subcommands.
991 Subcommand create creates pool.
993 Usage:
995 ceph osd pool create <poolname> {<int[0-]>} {<int[0-]>} {replicated|erasure}
997 {<erasure_code_profile>} {<rule>} {<int>} {--autoscale-mode=<on,off,warn>}
998 Subcommand delete deletes pool.
1000 Usage:
1002 ceph osd pool delete <poolname> {<poolname>} {--yes-i-really-really-mean-it}
1004 Subcommand get gets pool parameter <var>.
1006 Usage:
1008 ceph osd pool get <poolname> size|min_size|pg_num|pgp_num|crush_rule|write_fadvise_dontneed
1010 Only for tiered pools:
1012 ceph osd pool get <poolname> hit_set_type|hit_set_period|hit_set_count|hit_set_fpp|
1014 target_max_objects|target_max_bytes|cache_target_dirty_ratio|cache_target_dirty_high_ratio|
1015 cache_target_full_ratio|cache_min_flush_age|cache_min_evict_age|
1016 min_read_recency_for_promote|hit_set_grade_decay_rate|hit_set_search_last_n
1017 Only for erasure coded pools:
1019 ceph osd pool get <poolname> erasure_code_profile
1021 Use all to get all pool parameters that apply to the pool's type:
1023 ceph osd pool get <poolname> all
1025 Subcommand get-quota obtains object or byte limits for pool.
1027 Usage:
1029 ceph osd pool get-quota <poolname>
1031 Subcommand ls list pools
1033 Usage:
1035 ceph osd pool ls {detail}
1037 Subcommand mksnap makes snapshot <snap> in <pool>.
1039 Usage:
1041 ceph osd pool mksnap <poolname> <snap>
1043 Subcommand rename renames <srcpool> to <destpool>.
1045 Usage:
1047 ceph osd pool rename <poolname> <poolname>
1049 Subcommand rmsnap removes snapshot <snap> from <pool>.
1051 Usage:
1053 ceph osd pool rmsnap <poolname> <snap>
1055 Subcommand set sets pool parameter <var> to <val>.
1057 Usage:
1059 ceph osd pool set <poolname> size|min_size|pg_num|
1061 pgp_num|crush_rule|hashpspool|nodelete|nopgchange|nosizechange|
1062 hit_set_type|hit_set_period|hit_set_count|hit_set_fpp|debug_fake_ec_pool|
1063 target_max_bytes|target_max_objects|cache_target_dirty_ratio|
1064 cache_target_dirty_high_ratio|
1065 cache_target_full_ratio|cache_min_flush_age|cache_min_evict_age|
1066 min_read_recency_for_promote|write_fadvise_dontneed|hit_set_grade_decay_rate|
1067 hit_set_search_last_n
1068 <val> {--yes-i-really-mean-it}
1069 Subcommand set-quota sets object or byte limit on pool.
1071 Usage:
1073 ceph osd pool set-quota <poolname> max_objects|max_bytes <val>
1075 Subcommand stats obtain stats from all pools, or from specified pool.
1077 Usage:
1079 ceph osd pool stats {<name>}
1081 Subcommand application is used for adding an annotation to the given
1083 pool. By default, the possible applications are object, block, and file
1084 storage (corresponding app-names are "rgw", "rbd", and "cephfs"). How‐
1085 ever, there might be other applications as well. Based on the applica‐
1086 tion, there may or may not be some processing conducted.
1087 Subcommand disable disables the given application on the given pool.
1089 Usage:
1091 ceph osd pool application disable <pool-name> <app> {--yes-i-really-mean-it}
1093 Subcommand enable adds an annotation to the given pool for the men‐
1095 tioned application.
1096 Usage:
1098 ceph osd pool application enable <pool-name> <app> {--yes-i-really-mean-it}
1100 Subcommand get displays the value for the given key that is associated
1102 with the given application of the given pool. Not passing the optional
1103 arguments would display all key-value pairs for all applications for
1104 all pools.
1105 Usage:
1107 ceph osd pool application get {<pool-name>} {<app>} {<key>}
1109 Subcommand rm removes the key-value pair for the given key in the given
1111 application of the given pool.
1112 Usage:
1114 ceph osd pool application rm <pool-name> <app> <key>
1116 Subcommand set associates or updates, if it already exists, a key-value
1118 pair with the given application for the given pool.
1119 Usage:
1121 ceph osd pool application set <pool-name> <app> <key> <value>
1123 Subcommand primary-affinity adjust osd primary-affinity from 0.0
1125 <=<weight> <= 1.0
1126 Usage:
1128 ceph osd primary-affinity <osdname (id|osd.id)> <float[0.0-1.0]>
1130 Subcommand primary-temp sets primary_temp mapping pgid:<id>|-1 (devel‐
1132 opers only).
1133 Usage:
1135 ceph osd primary-temp <pgid> <id>
1137 Subcommand repair initiates repair on a specified osd.
1139 Usage:
1141 ceph osd repair <who>
1143 Subcommand reweight reweights osd to 0.0 < <weight> < 1.0.
1145 Usage:
1147 osd reweight <int[0-]> <float[0.0-1.0]>
1149 Subcommand reweight-by-pg reweight OSDs by PG distribution [over‐
1151 load-percentage-for-consideration, default 120].
1152 Usage:
1154 ceph osd reweight-by-pg {<int[100-]>} {<poolname> [<poolname...]}
1156 {--no-increasing}
1157 Subcommand reweight-by-utilization reweights OSDs by utilization. It
1159 only reweights outlier OSDs whose utilization exceeds the average, eg.
1160 the default 120% limits reweight to those OSDs that are more than 20%
1161 over the average. [overload-threshold, default 120 [max_weight_change,
1162 default 0.05 [max_osds_to_adjust, default 4]]]
1163 Usage:
1165 ceph osd reweight-by-utilization {<int[100-]> {<float[0.0-]> {<int[0-]>}}}
1167 {--no-increasing}
1168 Subcommand rm removes osd(s) <id> [<id>...] from the OSD map.
1170 Usage:
1172 ceph osd rm <ids> [<ids>...]
1174 Subcommand destroy marks OSD id as destroyed, removing its cephx en‐
1176 tity's keys and all of its dm-crypt and daemon-private config key en‐
1177 tries.
1178 This command will not remove the OSD from crush, nor will it remove the
1180 OSD from the OSD map. Instead, once the command successfully completes,
1181 the OSD will show marked as destroyed.
1182 In order to mark an OSD as destroyed, the OSD must first be marked as
1184 lost.
1185 Usage:
1187 ceph osd destroy <id> {--yes-i-really-mean-it}
1189 Subcommand purge performs a combination of osd destroy, osd rm and osd
1191 crush remove.
1192 Usage:
1194 ceph osd purge <id> {--yes-i-really-mean-it}
1196 Subcommand safe-to-destroy checks whether it is safe to remove or de‐
1198 stroy an OSD without reducing overall data redundancy or durability.
1199 It will return a success code if it is definitely safe, or an error
1200 code and informative message if it is not or if no conclusion can be
1201 drawn at the current time.
1202 Usage:
1204 ceph osd safe-to-destroy <id> [<ids>...]
1206 Subcommand scrub initiates scrub on specified osd.
1208 Usage:
1210 ceph osd scrub <who>
1212 Subcommand set sets cluster-wide <flag> by updating OSD map. The full
1214 flag is not honored anymore since the Mimic release, and ceph osd set
1215 full is not supported in the Octopus release.
1216 Usage:
1218 ceph osd set pause|noup|nodown|noout|noin|nobackfill|
1220 norebalance|norecover|noscrub|nodeep-scrub|notieragent
1221 Subcommand setcrushmap sets crush map from input file.
1223 Usage:
1225 ceph osd setcrushmap
1227 Subcommand setmaxosd sets new maximum osd value.
1229 Usage:
1231 ceph osd setmaxosd <int[0-]>
1233 Subcommand set-require-min-compat-client enforces the cluster to be
1235 backward compatible with the specified client version. This subcommand
1236 prevents you from making any changes (e.g., crush tunables, or using
1237 new features) that would violate the current setting. Please note, This
1238 subcommand will fail if any connected daemon or client is not compati‐
1239 ble with the features offered by the given <version>. To see the fea‐
1240 tures and releases of all clients connected to cluster, please see ceph
1241 features.
1242 Usage:
1244 ceph osd set-require-min-compat-client <version>
1246 Subcommand stat prints summary of OSD map.
1248 Usage:
1250 ceph osd stat
1252 Subcommand tier is used for managing tiers. It uses some additional
1254 subcommands.
1255 Subcommand add adds the tier <tierpool> (the second one) to base pool
1257 <pool> (the first one).
1258 Usage:
1260 ceph osd tier add <poolname> <poolname> {--force-nonempty}
1262 Subcommand add-cache adds a cache <tierpool> (the second one) of size
1264 <size> to existing pool <pool> (the first one).
1265 Usage:
1267 ceph osd tier add-cache <poolname> <poolname> <int[0-]>
1269 Subcommand cache-mode specifies the caching mode for cache tier <pool>.
1271 Usage:
1273 ceph osd tier cache-mode <poolname> writeback|readproxy|readonly|none
1275 Subcommand remove removes the tier <tierpool> (the second one) from
1277 base pool <pool> (the first one).
1278 Usage:
1280 ceph osd tier remove <poolname> <poolname>
1282 Subcommand remove-overlay removes the overlay pool for base pool
1284 <pool>.
1285 Usage:
1287 ceph osd tier remove-overlay <poolname>
1289 Subcommand set-overlay set the overlay pool for base pool <pool> to be
1291 <overlaypool>.
1292 Usage:
1294 ceph osd tier set-overlay <poolname> <poolname>
1296 Subcommand tree prints OSD tree.
1298 Usage:
1300 ceph osd tree {<int[0-]>}
1302 Subcommand unpause unpauses osd.
1304 Usage:
1306 ceph osd unpause
1308 Subcommand unset unsets cluster-wide <flag> by updating OSD map.
1310 Usage:
1312 ceph osd unset pause|noup|nodown|noout|noin|nobackfill|
1314 norebalance|norecover|noscrub|nodeep-scrub|notieragent
1315 pg
1317 It is used for managing the placement groups in OSDs. It uses some ad‐
1318 ditional subcommands.
1319 Subcommand debug shows debug info about pgs.
1321 Usage:
1323 ceph pg debug unfound_objects_exist|degraded_pgs_exist
1325 Subcommand deep-scrub starts deep-scrub on <pgid>.
1327 Usage:
1329 ceph pg deep-scrub <pgid>
1331 Subcommand dump shows human-readable versions of pg map (only 'all'
1333 valid with plain).
1334 Usage:
1336 ceph pg dump {all|summary|sum|delta|pools|osds|pgs|pgs_brief} [{all|summary|sum|delta|pools|osds|pgs|pgs_brief...]}
1338 Subcommand dump_json shows human-readable version of pg map in json
1340 only.
1341 Usage:
1343 ceph pg dump_json {all|summary|sum|delta|pools|osds|pgs|pgs_brief} [{all|summary|sum|delta|pools|osds|pgs|pgs_brief...]}
1345 Subcommand dump_pools_json shows pg pools info in json only.
1347 Usage:
1349 ceph pg dump_pools_json
1351 Subcommand dump_stuck shows information about stuck pgs.
1353 Usage:
1355 ceph pg dump_stuck {inactive|unclean|stale|undersized|degraded [inactive|unclean|stale|undersized|degraded...]}
1357 {<int>}
1358 Subcommand getmap gets binary pg map to -o/stdout.
1360 Usage:
1362 ceph pg getmap
1364 Subcommand ls lists pg with specific pool, osd, state
1366 Usage:
1368 ceph pg ls {<int>} {<pg-state> [<pg-state>...]}
1370 Subcommand ls-by-osd lists pg on osd [osd]
1372 Usage:
1374 ceph pg ls-by-osd <osdname (id|osd.id)> {<int>}
1376 {<pg-state> [<pg-state>...]}
1377 Subcommand ls-by-pool lists pg with pool = [poolname]
1379 Usage:
1381 ceph pg ls-by-pool <poolstr> {<int>} {<pg-state> [<pg-state>...]}
1383 Subcommand ls-by-primary lists pg with primary = [osd]
1385 Usage:
1387 ceph pg ls-by-primary <osdname (id|osd.id)> {<int>}
1389 {<pg-state> [<pg-state>...]}
1390 Subcommand map shows mapping of pg to osds.
1392 Usage:
1394 ceph pg map <pgid>
1396 Subcommand repair starts repair on <pgid>.
1398 Usage:
1400 ceph pg repair <pgid>
1402 Subcommand scrub starts scrub on <pgid>.
1404 Usage:
1406 ceph pg scrub <pgid>
1408 Subcommand stat shows placement group status.
1410 Usage:
1412 ceph pg stat
1414 quorum
1416 Cause a specific MON to enter or exit quorum.
1417 Usage:
1419 ceph tell mon.<id> quorum enter|exit
1421 quorum_status
1423 Reports status of monitor quorum.
1424 Usage:
1426 ceph quorum_status
1428 report
1430 Reports full status of cluster, optional title tag strings.
1431 Usage:
1433 ceph report {<tags> [<tags>...]}
1435 status
1437 Shows cluster status.
1438 Usage:
1440 ceph status
1442 tell
1444 Sends a command to a specific daemon.
1445 Usage:
1447 ceph tell <name (type.id)> <command> [options...]
1449 List all available commands.
1451 Usage:
1453 ceph tell <name (type.id)> help
1455 version
1457 Show mon daemon version
1458 Usage:
1460 ceph version
1462
1464 -i infile
1465 will specify an input file to be passed along as a payload with
1466 the command to the monitor cluster. This is only used for spe‐
1467 cific monitor commands.
1468 -o outfile
1470 will write any payload returned by the monitor cluster with its
1471 reply to outfile. Only specific monitor commands (e.g. osd
1472 getmap) return a payload.
1473 --setuser user
1475 will apply the appropriate user ownership to the file specified
1476 by the option '-o'.
1477 --setgroup group
1479 will apply the appropriate group ownership to the file specified
1480 by the option '-o'.
1481 -c ceph.conf, --conf=ceph.conf
1483 Use ceph.conf configuration file instead of the default
1484 /etc/ceph/ceph.conf to determine monitor addresses during
1485 startup.
1486 --id CLIENT_ID, --user CLIENT_ID
1488 Client id for authentication.
1489 --name CLIENT_NAME, -n CLIENT_NAME
1491 Client name for authentication.
1492 --cluster CLUSTER
1494 Name of the Ceph cluster.
1495 --admin-daemon ADMIN_SOCKET, daemon DAEMON_NAME
1497 Submit admin-socket commands via admin sockets in /var/run/ceph.
1498 --admin-socket ADMIN_SOCKET_NOPE
1500 You probably mean --admin-daemon
1501 -s, --status
1503 Show cluster status.
1504 -w, --watch
1506 Watch live cluster changes on the default 'cluster' channel
1507 -W, --watch-channel
1509 Watch live cluster changes on any channel (cluster, audit,
1510 cephadm, or * for all)
1511 --watch-debug
1513 Watch debug events.
1514 --watch-info
1516 Watch info events.
1517 --watch-sec
1519 Watch security events.
1520 --watch-warn
1522 Watch warning events.
1523 --watch-error
1525 Watch error events.
1526 --version, -v
1528 Display version.
1529 --verbose
1531 Make verbose.
1532 --concise
1534 Make less verbose.
1535 -f {json,json-pretty,xml,xml-pretty,plain,yaml}, --format
1537 Format of output. Note: yaml is only valid for orch commands.
1538 --connect-timeout CLUSTER_TIMEOUT
1540 Set a timeout for connecting to the cluster.
1541 --no-increasing
1543 --no-increasing is off by default. So increasing the osd weight
1544 is allowed using the reweight-by-utilization or
1545 test-reweight-by-utilization commands. If this option is used
1546 with these commands, it will help not to increase osd weight
1547 even the osd is under utilized.
1548 --block
1550 block until completion (scrub and deep-scrub only)
1551
1553 ceph is part of Ceph, a massively scalable, open-source, distributed
1554 storage system. Please refer to the Ceph documentation at
1555 https://docs.ceph.com for more information.
1556
1558 ceph-mon(8), ceph-osd(8), ceph-mds(8)
1559
1561 2010-2022, Inktank Storage, Inc. and contributors. Licensed under Cre‐
1562 ative Commons Attribution Share Alike 3.0 (CC-BY-SA-3.0)
1563dev Oct 18, 2022 CEPH(8)