cephadm-adopt.yml 40.4 KB
Newer Older
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
---
#
# This playbook does a cephadm adopt for all the Ceph services
#

- name: confirm whether user really meant to adopt the cluster by cephadm
  hosts: localhost
  connection: local
  become: false
  gather_facts: false
  vars_prompt:
    - name: ireallymeanit
      prompt: Are you sure you want to adopt the cluster by cephadm ?
      default: 'no'
      private: no
  tasks:
    - name: exit playbook, if user did not mean to adopt the cluster by cephadm
      fail:
        msg: >
          Exiting cephadm-adopt playbook, cluster was NOT adopted.
           To adopt the cluster, either say 'yes' on the prompt or
           use `-e ireallymeanit=yes` on the command line when
           invoking the playbook
      when: ireallymeanit != 'yes'

- name: gather facts and prepare system for cephadm
  hosts:
    - "{{ mon_group_name|default('mons') }}"
    - "{{ osd_group_name|default('osds') }}"
    - "{{ mds_group_name|default('mdss') }}"
    - "{{ rgw_group_name|default('rgws') }}"
    - "{{ mgr_group_name|default('mgrs') }}"
    - "{{ rbdmirror_group_name|default('rbdmirrors') }}"
    - "{{ nfs_group_name|default('nfss') }}"
    - "{{ iscsi_gw_group_name|default('iscsigws') }}"
    - "{{ grafana_server_group_name|default('grafana-server') }}"
  become: true
  gather_facts: false
  vars:
    delegate_facts_host: true
  tasks:
    - import_role:
        name: ceph-defaults

    - name: gather facts
      setup:
47
48
49
50
        gather_subset:
          - 'all'
          - '!facter'
          - '!ohai'
51
52
53
54
      when: not delegate_facts_host | bool or inventory_hostname in groups.get(client_group_name, [])

    - name: gather and delegate facts
      setup:
55
56
57
58
        gather_subset:
          - 'all'
          - '!facter'
          - '!ohai'
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
      delegate_to: "{{ item }}"
      delegate_facts: true
      with_items: "{{ groups['all'] | difference(groups.get('clients', [])) }}"
      run_once: true
      when: delegate_facts_host | bool

    - name: fail if one osd node is using filestore
      fail:
        msg: >
          filestore OSDs are not supported with cephadm.
          Please convert them with the filestore-to-bluestore.yml playbook first.
      when:
        - osd_group_name in group_names
        - osd_objectstore == 'filestore'

    - import_role:
        name: ceph-facts
        tasks_from: container_binary.yml

    - name: get the ceph version
      command: "{{ container_binary + ' run --rm --entrypoint=ceph ' + ceph_docker_registry + '/' + ceph_docker_image + ':' + ceph_docker_image_tag if containerized_deployment | bool else 'ceph' }} --version"
      changed_when: false
      register: ceph_version_out

    - name: set_fact ceph_version
      set_fact:
        ceph_version: "{{ ceph_version_out.stdout.split(' ')[2] }}"

    - name: fail on pre octopus ceph releases
      fail:
        msg: >
          Your Ceph version {{ ceph_version }} is not supported for this operation.
          Please upgrade your cluster with the rolling_update.yml playbook first.
      when: ceph_version is version('15.2', '<')

    - name: check if it is atomic host
      stat:
        path: /run/ostree-booted
      register: stat_ostree

    - name: set_fact is_atomic
      set_fact:
        is_atomic: "{{ stat_ostree.stat.exists }}"

    - import_role:
        name: ceph-container-engine
      when: not containerized_deployment | bool

    - import_role:
        name: ceph-container-common
        tasks_from: registry.yml
      when:
        - not containerized_deployment | bool
        - ceph_docker_registry_auth | bool

    - name: "pulling {{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }} image"
      command: "{{ timeout_command }} {{ container_binary }} pull {{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}"
      changed_when: false
      register: docker_image
      until: docker_image.rc == 0
      retries: "{{ docker_pull_retry }}"
      delay: 10
      when:
        - not containerized_deployment | bool
        - inventory_hostname in groups.get(mon_group_name, []) or
          inventory_hostname in groups.get(osd_group_name, []) or
          inventory_hostname in groups.get(mds_group_name, []) or
          inventory_hostname in groups.get(rgw_group_name, []) or
          inventory_hostname in groups.get(mgr_group_name, []) or
          inventory_hostname in groups.get(rbdmirror_group_name, []) or
          inventory_hostname in groups.get(iscsi_gw_group_name, []) or
          inventory_hostname in groups.get(nfs_group_name, [])

    - name: install cephadm requirements
      package:
        name: ['python3', 'lvm2']
      register: result
      until: result is succeeded

    - name: install cephadm
      package:
        name: cephadm
      register: result
      until: result is succeeded
      when: not containerized_deployment | bool

    - name: install cephadm mgr module
      package:
        name: ceph-mgr-cephadm
      register: result
      until: result is succeeded
      when:
        - not containerized_deployment | bool
        - mgr_group_name in group_names

    - name: get cephadm from the container image
      when: containerized_deployment | bool
      block:
        - name: create a cephadm container
          command: "{{ container_binary }} create --name cephadm {{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}"
          changed_when: false

        - name: cp the cephadm cli file
          command: "{{ container_binary }} cp cephadm:/usr/sbin/cephadm /usr/sbin/cephadm"
          args:
            creates: /usr/sbin/cephadm

        - name: remove the cephadm container
          command: "{{ container_binary }} rm cephadm"
          changed_when: false

    - name: set_fact container_exec_cmd
      set_fact:
        container_exec_cmd: "{{ container_binary }} exec ceph-mon-{{ hostvars[groups[mon_group_name][0]]['ansible_hostname'] }}"
      when: containerized_deployment | bool

    - name: get current fsid
      command: "{{ container_exec_cmd | default('') }} ceph --admin-daemon /var/run/ceph/{{ cluster }}-mon.{{ hostvars[groups[mon_group_name][0]]['ansible_hostname'] }}.asok config get fsid --format json"
      register: current_fsid
      run_once: true
      changed_when: false
      delegate_to: "{{ groups[mon_group_name][0] }}"

    - name: set_fact fsid
      set_fact:
        fsid: "{{ (current_fsid.stdout | from_json).fsid }}"

    - name: enable cephadm mgr module
      command: "{{ container_exec_cmd | default('') }} ceph --cluster {{ cluster }} mgr module enable cephadm"
      changed_when: false
      run_once: true
      delegate_to: '{{ groups[mon_group_name][0] }}'

    - name: set cephadm as orchestrator backend
      command: "{{ container_exec_cmd | default('') }} ceph --cluster {{ cluster }} orch set backend cephadm"
      changed_when: false
      run_once: true
      delegate_to: '{{ groups[mon_group_name][0] }}'

    - name: generate cephadm ssh key
      command: "{{ container_exec_cmd | default('') }} ceph --cluster {{ cluster }} cephadm generate-key"
      changed_when: false
      run_once: true
      delegate_to: '{{ groups[mon_group_name][0] }}'

    - name: get the cephadm ssh pub key
      command: "{{ container_exec_cmd | default('') }} ceph --cluster {{ cluster }} cephadm get-pub-key"
      changed_when: false
      run_once: true
      register: cephadm_pubpkey
      delegate_to: '{{ groups[mon_group_name][0] }}'

    - name: allow cephadm key for root account
      authorized_key:
        user: root
        key: '{{ cephadm_pubpkey.stdout }}'

    - name: run cephadm prepare-host
      command: cephadm prepare-host
      changed_when: false
      environment:
        CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

    - name: set default container image in ceph configuration
      command: "{{ container_exec_cmd | default('') }} ceph --cluster {{ cluster }} config set global container_image {{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}"
      changed_when: false
      run_once: true
      delegate_to: '{{ groups[mon_group_name][0] }}'

228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
    - name: set container image base in ceph configuration
      command: "{{ container_exec_cmd | default('') }} ceph --cluster {{ cluster }} config set mgr mgr/cephadm/container_image_base {{ ceph_docker_registry }}/{{ ceph_docker_image }}"
      changed_when: false
      run_once: true
      delegate_to: '{{ groups[mon_group_name][0] }}'

    - name: set dashboard container image in ceph mgr configuration
      when: dashboard_enabled | bool
      run_once: true
      block:
        - name: set alertmanager container image in ceph configuration
          command: "{{ container_exec_cmd | default('') }} ceph --cluster {{ cluster }} config set mgr mgr/cephadm/container_image_alertmanager {{ alertmanager_container_image }}"
          changed_when: false
          delegate_to: '{{ groups[mon_group_name][0] }}'

        - name: set grafana container image in ceph configuration
          command: "{{ container_exec_cmd | default('') }} ceph --cluster {{ cluster }} config set mgr mgr/cephadm/container_image_grafana {{ grafana_container_image }}"
          changed_when: false
          delegate_to: '{{ groups[mon_group_name][0] }}'

        - name: set node-exporter container image in ceph configuration
          command: "{{ container_exec_cmd | default('') }} ceph --cluster {{ cluster }} config set mgr mgr/cephadm/container_image_node_exporter {{ node_exporter_container_image }}"
          changed_when: false
          delegate_to: '{{ groups[mon_group_name][0] }}'

        - name: set prometheus container image in ceph configuration
          command: "{{ container_exec_cmd | default('') }} ceph --cluster {{ cluster }} config set mgr mgr/cephadm/container_image_prometheus {{ prometheus_container_image }}"
          changed_when: false
          delegate_to: '{{ groups[mon_group_name][0] }}'

258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
    - name: manage nodes with cephadm
      command: "{{ container_exec_cmd | default('') }} ceph --cluster {{ cluster }} orch host add {{ hostvars[item]['ansible_hostname'] }} {{ hostvars[item]['ansible_default_ipv4']['address'] }} {{ hostvars[item]['group_names'] | join(' ') }}"
      changed_when: false
      run_once: true
      loop: '{{ ansible_play_hosts_all }}'
      delegate_to: '{{ groups[mon_group_name][0] }}'

    - name: add ceph label for core component
      command: "{{ container_exec_cmd | default('') }} ceph --cluster {{ cluster }} orch host label add {{ hostvars[item]['ansible_hostname'] }} ceph"
      changed_when: false
      run_once: true
      loop: '{{ ansible_play_hosts_all }}'
      delegate_to: '{{ groups[mon_group_name][0] }}'
      when: item in groups.get(mon_group_name, []) or
            item in groups.get(osd_group_name, []) or
            item in groups.get(mds_group_name, []) or
            item in groups.get(rgw_group_name, []) or
            item in groups.get(mgr_group_name, []) or
            item in groups.get(rbdmirror_group_name, [])

    - name: set_fact ceph_cmd
      set_fact:
        ceph_cmd: "{{ container_binary + ' run --rm --net=host -v /etc/ceph:/etc/ceph:z -v /var/lib/ceph:/var/lib/ceph:z -v /var/run/ceph:/var/run/ceph:z --entrypoint=ceph ' + ceph_docker_registry + '/' + ceph_docker_image + ':' + ceph_docker_image_tag if containerized_deployment | bool else 'ceph' }}"

282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
    - name: get the client.admin keyring
      command: "{{ ceph_cmd }} --cluster {{ cluster }} auth get client.admin"
      changed_when: false
      run_once: true
      delegate_to: '{{ groups[mon_group_name][0] }}'
      register: client_admin_keyring

    - name: copy the client.admin keyring
      copy:
        dest: "/etc/ceph/{{ cluster }}.client.admin.keyring"
        content: "{{ client_admin_keyring.stdout + '\n' }}"
        owner: "{{ ceph_uid | int if containerized_deployment | bool else 'ceph' }}"
        group: "{{ ceph_uid | int if containerized_deployment | bool else 'ceph' }}"
        mode: "{{ ceph_keyring_permissions }}"
      run_once: true
      delegate_to: "{{ item }}"
      with_items:
        - "{{ groups.get(osd_group_name, []) }}"
        - "{{ groups.get(mds_group_name, []) }}"
        - "{{ groups.get(rgw_group_name, []) }}"
        - "{{ groups.get(mgr_group_name, []) }}"
        - "{{ groups.get(rbdmirror_group_name, []) }}"

305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
    - name: assimilate ceph configuration
      command: "{{ ceph_cmd }} --cluster {{ cluster }} config assimilate-conf -i /etc/ceph/{{ cluster }}.conf"
      changed_when: false
      when: inventory_hostname in groups.get(mon_group_name, []) or
            inventory_hostname in groups.get(osd_group_name, []) or
            inventory_hostname in groups.get(mds_group_name, []) or
            inventory_hostname in groups.get(rgw_group_name, []) or
            inventory_hostname in groups.get(mgr_group_name, []) or
            inventory_hostname in groups.get(rbdmirror_group_name, [])

- name: adopt ceph mon daemons
  hosts: "{{ mon_group_name|default('mons') }}"
  serial: 1
  become: true
  gather_facts: false
  tasks:
    - import_role:
        name: ceph-defaults

    - name: adopt mon daemon
      command: "cephadm adopt --cluster {{ cluster }} --skip-pull --style legacy --name mon.{{ ansible_hostname }} {{ '--skip-firewalld' if not configure_firewall | bool else '' }} {{ '--docker' if container_binary == 'docker' else '' }}"
      args:
        creates: '/var/lib/ceph/{{ fsid }}/mon.{{ ansible_hostname }}/unit.run'
      environment:
        CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

    - name: reset failed ceph-mon systemd unit
      command: 'systemctl reset-failed ceph-mon@{{ ansible_hostname }}'
      changed_when: false
334
      failed_when: false
335
336
337
338
339
340
341
342
343
344
345
346
347
348
      when: containerized_deployment | bool

    - name: remove ceph-mon systemd unit file
      file:
        path: /etc/systemd/system/ceph-mon@.service
        state: absent
      when: containerized_deployment | bool

    - name: remove ceph-mon systemd override directory
      file:
        path: /etc/systemd/system/ceph-mon@.service.d
        state: absent
      when: not containerized_deployment | bool

349
350
351
352
353
354
355
356
357
358
359
    - name: waiting for the monitor to join the quorum...
      command: "cephadm shell --fsid {{ fsid }} -- ceph --cluster {{ cluster }} -s --format json"
      changed_when: false
      register: ceph_health_raw
      until: >
        ansible_hostname in (ceph_health_raw.stdout | from_json)["quorum_names"]
      retries: "{{ health_mon_check_retries }}"
      delay: "{{ health_mon_check_delay }}"
      environment:
        CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
- name: adopt ceph mgr daemons
  hosts: "{{ mgr_group_name|default('mgrs') }}"
  serial: 1
  become: true
  gather_facts: false
  tasks:
    - import_role:
        name: ceph-defaults

    - name: adopt mgr daemon
      command: "cephadm adopt --cluster {{ cluster }} --skip-pull --style legacy --name mgr.{{ ansible_hostname }} {{ '--skip-firewalld' if not configure_firewall | bool else '' }} {{ '--docker' if container_binary == 'docker' else '' }}"
      args:
        creates: '/var/lib/ceph/{{ fsid }}/mgr.{{ ansible_hostname }}/unit.run'
      environment:
        CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

    - name: reset failed ceph-mgr systemd unit
      command: 'systemctl reset-failed ceph-mgr@{{ ansible_hostname }}'
      changed_when: false
379
      failed_when: false
380
381
382
383
384
385
386
387
388
389
390
391
392
393
      when: containerized_deployment | bool

    - name: remove ceph-mgr systemd unit file
      file:
        path: /etc/systemd/system/ceph-mgr@.service
        state: absent
      when: containerized_deployment | bool

    - name: remove ceph-mgr systemd override directory
      file:
        path: /etc/systemd/system/ceph-mgr@.service.d
        state: absent
      when: not containerized_deployment | bool

394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
- name: set osd flags
  hosts: "{{ mon_group_name|default('mons') }}"
  become: true
  gather_facts: false
  tasks:
    - import_role:
        name: ceph-defaults

    - name: set osd flags
      command: "cephadm shell --fsid {{ fsid }} -- ceph --cluster {{ cluster }} osd set {{ item }}"
      run_once: true
      changed_when: false
      with_items:
        - noout
        - nodeep-scrub
      environment:
        CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
- name: adopt ceph osd daemons
  hosts: "{{ osd_group_name|default('osd') }}"
  serial: 1
  become: true
  gather_facts: false
  tasks:
    - import_role:
        name: ceph-defaults

    - import_role:
        name: ceph-facts
        tasks_from: container_binary.yml
      when: containerized_deployment | bool

    - name: set_fact ceph_volume
      set_fact:
        ceph_volume: "{{ container_binary + ' run --rm --privileged=true --net=host --pid=host --ipc=host -v /dev:/dev -v /etc/ceph:/etc/ceph:z -v /var/lib/ceph:/var/lib/ceph:z -v /var/run:/var/run --entrypoint=ceph-volume ' + ceph_docker_registry + '/' + ceph_docker_image + ':' + ceph_docker_image_tag if containerized_deployment | bool else 'ceph-volume' }}"

    - name: get osd list
      command: "{{ ceph_volume }} --cluster {{ cluster }} lvm list --format json"
      changed_when: false
      register: osd_list

    - name: set osd fsid for containerized deployment
      lineinfile:
        path: '/var/lib/ceph/osd/{{ cluster }}-{{ item.key }}/fsid'
        line: "{{ (item.value | selectattr('type', 'equalto', 'block') | map(attribute='tags') | first)['ceph.osd_fsid'] }}"
        owner: '{{ ceph_uid }}'
        group: '{{ ceph_uid }}'
        create: true
      with_dict: '{{ osd_list.stdout | from_json }}'
      when: containerized_deployment | bool

    - name: set osd type for containerized deployment
      lineinfile:
        path: '/var/lib/ceph/osd/{{ cluster }}-{{ item }}/type'
        line: 'bluestore'
        owner: '{{ ceph_uid }}'
        group: '{{ ceph_uid }}'
        create: true
      loop: '{{ (osd_list.stdout | from_json).keys() | list }}'
      when: containerized_deployment | bool

    - name: adopt osd daemon
      command: "cephadm adopt --cluster {{ cluster }} --skip-pull --style legacy --name osd.{{ item }} {{ '--skip-firewalld' if not configure_firewall | bool else '' }} {{ '--docker' if container_binary == 'docker' else '' }}"
      loop: '{{ (osd_list.stdout | from_json).keys() | list }}'
      args:
        creates: '/var/lib/ceph/{{ fsid }}/osd.{{ item }}/unit.run'
      environment:
        CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

    - name: remove ceph-osd systemd unit and ceph-osd-run.sh files
      file:
        path: '{{ item }}'
        state: absent
      loop:
        - /etc/systemd/system/ceph-osd@.service
        - "{{ ceph_osd_docker_run_script_path | default('/usr/share') }}/ceph-osd-run.sh"
      when: containerized_deployment | bool

    - name: remove ceph-osd systemd override directory
      file:
        path: /etc/systemd/system/ceph-osd@.service.d
        state: absent
      when: not containerized_deployment | bool

478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
    - name: waiting for clean pgs...
      command: "cephadm shell --fsid {{ fsid }} -- ceph --cluster {{ cluster }} -s --format json"
      changed_when: false
      register: ceph_health_post
      until: >
        (((ceph_health_post.stdout | from_json).pgmap.pgs_by_state | length) > 0)
        and
        (((ceph_health_post.stdout | from_json).pgmap.pgs_by_state | selectattr('state_name', 'search', '^active\\+clean') | map(attribute='count') | list | sum) == (ceph_health_post.stdout | from_json).pgmap.num_pgs)
      delegate_to: "{{ groups[mon_group_name][0] }}"
      retries: "{{ health_osd_check_retries }}"
      delay: "{{ health_osd_check_delay }}"
      environment:
        CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

- name: unset osd flags
  hosts: "{{ mon_group_name|default('mons') }}"
  become: true
  gather_facts: false
  tasks:
    - import_role:
        name: ceph-defaults

    - name: unset osd flags
      command: "cephadm shell --fsid {{ fsid }} -- ceph --cluster {{ cluster }} osd unset {{ item }}"
      run_once: true
      changed_when: false
      with_items:
        - noout
        - nodeep-scrub
      environment:
        CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

510
- name: redeploy mds daemons
511
  hosts: "{{ mon_group_name|default('mons') }}"
512
513
514
515
516
517
518
  become: true
  gather_facts: false
  tasks:
    - import_role:
        name: ceph-defaults

    - name: update the placement of metadata hosts
519
      command: "cephadm shell --fsid {{ fsid }} -- ceph --cluster {{ cluster }} orch apply mds {{ cephfs }} --placement='{{ groups.get(mds_group_name, []) | length }} label:{{ mds_group_name }}'"
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
      run_once: true
      changed_when: false
      environment:
        CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

- name: stop and remove legacy ceph mds daemons
  hosts: "{{ mds_group_name|default('mdss') }}"
  serial: 1
  become: true
  gather_facts: false
  tasks:
    - import_role:
        name: ceph-defaults

    - name: stop and disable ceph-mds systemd service
      service:
        name: 'ceph-mds@{{ ansible_hostname }}'
        state: stopped
        enabled: false

    - name: stop and disable ceph-mds systemd target
      service:
        name: ceph-mds.target
        state: stopped
        enabled: false
      when: not containerized_deployment | bool

    - name: reset failed ceph-mds systemd unit
      command: 'systemctl reset-failed ceph-mds@{{ ansible_hostname }}'
      changed_when: false
550
      failed_when: false
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
      when: containerized_deployment | bool

    - name: remove ceph-mds systemd unit file
      file:
        path: /etc/systemd/system/ceph-mds@.service
        state: absent
      when: containerized_deployment | bool

    - name: remove ceph-mds systemd override directory
      file:
        path: /etc/systemd/system/ceph-mds@.service.d
        state: absent
      when: not containerized_deployment | bool

    - name: remove legacy ceph mds data
      file:
        path: '/var/lib/ceph/mds/{{ cluster }}-{{ ansible_hostname }}'
        state: absent

- name: rgw realm/zonegroup/zone requirements
571
  hosts: "{{ mon_group_name|default('mons') }}"
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
  become: true
  gather_facts: false
  tasks:
    - import_role:
        name: ceph-defaults

    - name: for non multisite setup
      when: not rgw_multisite | bool
      run_once: true
      block:
        - name: create a default realm
          command: "cephadm shell --fsid {{ fsid }} -- radosgw-admin --cluster {{ cluster }} realm create --rgw-realm=default --default"
          run_once: true
          changed_when: false
          environment:
            CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

        - name: modify the default zonegroup
          command: "cephadm shell --fsid {{ fsid }} -- radosgw-admin --cluster {{ cluster }} zonegroup modify --rgw-realm=default --rgw-zonegroup=default"
          run_once: true
          changed_when: false
          environment:
            CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

        - name: modify the default zone
          command: "cephadm shell --fsid {{ fsid }} -- radosgw-admin --cluster {{ cluster }} zone modify --rgw-realm=default --rgw-zonegroup=default --rgw-zone=default"
          run_once: true
          changed_when: false
          environment:
            CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

        - name: commit the period
          command: "cephadm shell --fsid {{ fsid }} -- radosgw-admin --cluster {{ cluster }} period update --commit"
          run_once: true
          changed_when: false
          environment:
            CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

610
    - name: update the placement of radosgw hosts
611
      command: "cephadm shell --fsid {{ fsid }} -- ceph --cluster {{ cluster }} orch apply rgw {{ rgw_realm | default('default') }} {{ rgw_zone | default('default') }} --placement='{{ groups.get(rgw_group_name, []) | length }} label:{{ rgw_group_name }}' --port {{ radosgw_frontend_port }} {{ '--ssl' if radosgw_frontend_ssl_certificate else '' }}"
612
613
614
615
616
      run_once: true
      changed_when: false
      environment:
        CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
- name: redeploy rgw daemons
  hosts: "{{ rgw_group_name|default('rgws') }}"
  serial: 1
  become: true
  gather_facts: false
  tasks:
    - import_role:
        name: ceph-defaults

    - import_role:
        name: ceph-facts
        tasks_from: set_radosgw_address.yml

    - name: stop and disable ceph-radosgw systemd service
      service:
        name: 'ceph-radosgw@rgw.{{ ansible_hostname }}.{{ item.instance_name }}'
        state: stopped
        enabled: false
      loop: '{{ rgw_instances }}'

    - name: stop and disable ceph-radosgw systemd target
      service:
        name: ceph-rgw.target
        state: stopped
        enabled: false
      when: not containerized_deployment | bool

    - name: reset failed ceph-radosgw systemd unit
      command: 'systemctl reset-failed ceph-radosgw@rgw.{{ ansible_hostname }}.{{ item.instance_name }}'
      changed_when: false
647
      failed_when: false
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
      loop: '{{ rgw_instances }}'
      when: containerized_deployment | bool

    - name: remove ceph-radosgw systemd unit file
      file:
        path: /etc/systemd/system/ceph-radosgw@.service
        state: absent
      when: containerized_deployment | bool

    - name: remove ceph-radosgw systemd override directory
      file:
        path: /etc/systemd/system/ceph-radosgw@.service.d
        state: absent
      when: not containerized_deployment | bool

    - name: remove legacy ceph radosgw data
      file:
        path: '/var/lib/ceph/radosgw/{{ cluster }}-rgw.{{ ansible_hostname }}.{{ item.instance_name }}'
        state: absent
      loop: '{{ rgw_instances }}'

    - name: remove legacy ceph radosgw directory
      file:
        path: '/var/lib/ceph/radosgw/{{ cluster }}-rgw.{{ ansible_hostname }}'
        state: absent

- name: redeploy rbd-mirror daemons
675
  hosts: "{{ mon_group_name|default('mons') }}"
676
677
678
679
680
681
682
  become: true
  gather_facts: false
  tasks:
    - import_role:
        name: ceph-defaults

    - name: update the placement of rbd-mirror hosts
683
      command: "cephadm shell --fsid {{ fsid }} -- ceph --cluster {{ cluster }} orch apply rbd-mirror --placement='{{ groups.get(rbdmirror_group_name, []) | length }} label:{{ rbdmirror_group_name }}'"
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
      run_once: true
      changed_when: false
      environment:
        CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

- name: stop and remove legacy rbd-mirror daemons
  hosts: "{{ rbdmirror_group_name|default('rbdmirrors') }}"
  serial: 1
  become: true
  gather_facts: false
  tasks:
    - import_role:
        name: ceph-defaults

    - name: stop and disable rbd-mirror systemd service
      service:
        name: 'ceph-rbd-mirror@rbd-mirror.{{ ansible_hostname }}'
        state: stopped
        enabled: false

    - name: stop and disable rbd-mirror systemd target
      service:
        name: ceph-rbd-mirror.target
        state: stopped
        enabled: false
      when: not containerized_deployment | bool

    - name: reset failed rbd-mirror systemd unit
      command: 'systemctl reset-failed ceph-rbd-mirror@rbd-mirror.{{ ansible_hostname }}'
      changed_when: false
714
      failed_when: false
715
716
717
718
719
720
721
722
723
724
725
726
727
728
      when: containerized_deployment | bool

    - name: remove rbd-mirror systemd unit file
      file:
        path: /etc/systemd/system/ceph-rbd-mirror@.service
        state: absent
      when: containerized_deployment | bool

    - name: remove rbd-mirror systemd override directory
      file:
        path: /etc/systemd/system/ceph-rbd-mirror@.service.d
        state: absent
      when: not containerized_deployment | bool

729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
- name: redeploy iscsigw daemons
  hosts: "{{ iscsi_gw_group_name|default('iscsigws') }}"
  become: true
  gather_facts: false
  tasks:
    - import_role:
        name: ceph-defaults

    - name: update the placement of iscsigw hosts
      command: "cephadm shell --fsid {{ fsid }} -- ceph --cluster {{ cluster }} orch apply iscsi {{ iscsi_pool_name | default('rbd') }} {{ api_user | default('admin') }} {{ api_password | default('admin') }} {{ trusted_ip_list | default('192.168.122.1') }} --placement='{{ groups.get(iscsi_gw_group_name, []) | length }} label:{{ iscsi_gw_group_name }}'"
      run_once: true
      changed_when: false
      delegate_to: '{{ groups[mon_group_name][0] }}'
      environment:
        CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

- name: stop and remove legacy iscsigw daemons
  hosts: "{{ iscsi_gw_group_name|default('iscsigws') }}"
  serial: 1
  become: true
  gather_facts: false
  tasks:
    - import_role:
        name: ceph-defaults

    - name: stop and disable iscsigw systemd services
      service:
        name: '{{ item }}'
        state: stopped
        enabled: false
      with_items:
        - rbd-target-api
        - rbd-target-gw
        - tcmu-runner

    - name: reset failed iscsigw systemd units
      command: 'systemctl reset-failed {{ item }}'
      changed_when: false
      failed_when: false
      with_items:
        - rbd-target-api
        - rbd-target-gw
        - tcmu-runner
      when: containerized_deployment | bool

    - name: remove iscsigw systemd unit files
      file:
        path: '/etc/systemd/system/{{ item }}.service'
        state: absent
      with_items:
        - rbd-target-api
        - rbd-target-gw
        - tcmu-runner
      when: containerized_deployment | bool

784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
- name: redeploy ceph-crash daemons
  hosts:
    - "{{ mon_group_name|default('mons') }}"
    - "{{ osd_group_name|default('osds') }}"
    - "{{ mds_group_name|default('mdss') }}"
    - "{{ rgw_group_name|default('rgws') }}"
    - "{{ mgr_group_name|default('mgrs') }}"
    - "{{ rbdmirror_group_name|default('rbdmirrors') }}"
  become: true
  gather_facts: false
  tasks:
    - import_role:
        name: ceph-defaults

    - name: stop and disable ceph-crash systemd service
      service:
        name: ceph-crash
        state: stopped
        enabled: false
      failed_when: false
      when: not containerized_deployment | bool

    - name: update the placement of ceph-crash hosts
807
      command: "cephadm shell --fsid {{ fsid }} -- ceph --cluster {{ cluster }} orch apply crash --placement='label:ceph'"
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
      run_once: true
      changed_when: false
      environment:
        CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

- name: redeploy alertmanager/grafana/prometheus daemons
  hosts: "{{ grafana_server_group_name|default('grafana-server') }}"
  serial: 1
  become: true
  gather_facts: false
  tasks:
    - import_role:
        name: ceph-defaults

    - name: with dashboard enabled
      when: dashboard_enabled | bool
      block:
        # (workaround) cephadm adopt alertmanager only stops prometheus-alertmanager systemd service
        - name: stop and disable alertmanager systemd unit
          service:
            name: alertmanager
            state: stopped
            enabled: false

        # (workaround) cephadm adopt alertmanager only uses /etc/prometheus/alertmanager.yml
        - name: create alertmanager config symlink
          file:
            path: /etc/prometheus/alertmanager.yml
            src: '{{ alertmanager_conf_dir }}/alertmanager.yml'
            state: link

        # (workaround) cephadm adopt alertmanager only uses /var/lib/prometheus/alertmanager/
        - name: create alertmanager data symlink
          file:
            path: '{{ prometheus_data_dir }}/alertmanager'
            src: '{{ alertmanager_data_dir }}'
            state: link

        - name: adopt alertmanager daemon
          command: "cephadm adopt --cluster {{ cluster }} --skip-pull --style legacy --name alertmanager.{{ ansible_hostname }} {{ '--skip-firewalld' if not configure_firewall | bool else '' }} {{ '--docker' if container_binary == 'docker' else '' }}"
          args:
            creates: '/var/lib/ceph/{{ fsid }}/alertmanager.{{ ansible_hostname }}/unit.run'
          environment:
            CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

        - name: remove alertmanager systemd unit file
          file:
            path: /etc/systemd/system/alertmanager.service
            state: absent

        - name: remove the legacy alertmanager data
          file:
            path: '{{ alertmanager_data_dir }}'
            state: absent

        - name: stop and disable prometheus systemd unit
          service:
            name: prometheus
            state: stopped
            enabled: false

        - name: remove alertmanager data symlink
          file:
            path: '{{ prometheus_data_dir }}/alertmanager'
            state: absent

        # (workaround) cephadm adopt prometheus only uses /var/lib/prometheus/metrics/
        - name: tmp copy the prometheus data
          copy:
            src: '{{ prometheus_data_dir }}/'
            dest: /var/lib/prom_metrics
            owner: 65534
            group: 65534
            remote_src: true

        # (workaround) cephadm adopt prometheus only uses /var/lib/prometheus/metrics/
        - name: restore the prometheus data
          copy:
            src: /var/lib/prom_metrics/
            dest: /var/lib/prometheus/metrics
            owner: 65534
            group: 65534
            remote_src: true

        - name: remove the tmp prometheus data copy
          file:
            path: /var/lib/prom_metrics
            state: absent

        # (workaround) https://tracker.ceph.com/issues/45120
        - name: create missing prometheus target directory
          file:
            path: '/var/lib/ceph/{{ fsid }}/prometheus.{{ ansible_hostname }}/etc/prometheus'
            state: directory
            owner: 65534
            group: 65534
            recurse: true

        - name: adopt prometheus daemon
          command: "cephadm adopt --cluster {{ cluster }} --skip-pull --style legacy --name prometheus.{{ ansible_hostname }} {{ '--skip-firewalld' if not configure_firewall | bool else '' }} {{ '--docker' if container_binary == 'docker' else '' }}"
          args:
            creates: '/var/lib/ceph/{{ fsid }}/prometheus.{{ ansible_hostname }}/unit.run'
          environment:
            CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

        - name: remove prometheus systemd unit file
          file:
            path: /etc/systemd/system/prometheus.service
            state: absent

        - name: remove the legacy prometheus data
          file:
            path: '{{ prometheus_data_dir }}'
            state: absent

        # (workaround) cephadm adopt grafana only stops grafana systemd service
        - name: stop and disable grafana systemd unit
          service:
            name: grafana-server
            state: stopped
            enabled: false

        - name: adopt grafana daemon
          command: "cephadm adopt --cluster {{ cluster }} --skip-pull --style legacy --name grafana.{{ ansible_hostname }} {{ '--skip-firewalld' if not configure_firewall | bool else '' }} {{ '--docker' if container_binary == 'docker' else '' }}"
          args:
            creates: '/var/lib/ceph/{{ fsid }}/grafana.{{ ansible_hostname }}/unit.run'
          environment:
            CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

        - name: remove grafana systemd unit file
          file:
            path: /etc/systemd/system/grafana-server.service
            state: absent

        - name: remove the legacy grafana data
          file:
            path: /var/lib/grafana
            state: absent

- name: redeploy node-exporter daemons
  hosts:
    - "{{ mon_group_name|default('mons') }}"
    - "{{ osd_group_name|default('osds') }}"
    - "{{ mds_group_name|default('mdss') }}"
    - "{{ rgw_group_name|default('rgws') }}"
    - "{{ mgr_group_name|default('mgrs') }}"
    - "{{ rbdmirror_group_name|default('rbdmirrors') }}"
    - "{{ nfs_group_name|default('nfss') }}"
    - "{{ iscsi_gw_group_name|default('iscsigws') }}"
    - "{{ grafana_server_group_name|default('grafana-server') }}"
  become: true
  gather_facts: false
  tasks:
    - import_role:
        name: ceph-defaults

    - name: with dashboard enabled
      when: dashboard_enabled | bool
      block:
        - name: stop and disable node-exporter systemd service
          service:
            name: node_exporter
            state: stopped
            enabled: false

        - name: remove node_exporter systemd unit file
          file:
            path: /etc/systemd/system/node_exporter.service
            state: absent

        - name: update the placement of node-exporter hosts
979
          command: "cephadm shell --fsid {{ fsid }} -- ceph --cluster {{ cluster }} orch apply node-exporter --placement='*'"
980
981
982
983
984
985
986
987
988
989
990
991
992
993
          run_once: true
          changed_when: false
          environment:
            CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

- name: adjust placement daemons
  hosts: "{{ mon_group_name|default('mons') }}"
  become: true
  gather_facts: false
  tasks:
    - import_role:
        name: ceph-defaults

    - name: update the placement of monitor hosts
994
      command: "cephadm shell --fsid {{ fsid }} -- ceph --cluster {{ cluster }} orch apply mon --placement='{{ groups.get(mon_group_name, []) | length }} label:{{ mon_group_name }}'"
995
996
997
998
999
1000
      run_once: true
      changed_when: false
      environment:
        CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

    - name: update the placement of manager hosts
1001
      command: "cephadm shell --fsid {{ fsid }} -- ceph --cluster {{ cluster }} orch apply mgr --placement='{{ groups.get(mgr_group_name, []) | length }} label:{{ mgr_group_name }}'"
1002
1003
1004
1005
1006
1007
1008
1009
1010
      run_once: true
      changed_when: false
      environment:
        CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

    - name: with dashboard enabled
      when: dashboard_enabled | bool
      block:
        - name: update the placement of alertmanager hosts
1011
          command: "cephadm shell --fsid {{ fsid }} -- ceph --cluster {{ cluster }} orch apply alertmanager --placement='{{ groups.get(grafana_server_group_name, []) | length }} label:{{ grafana_server_group_name }}'"
1012
1013
1014
1015
1016
1017
          run_once: true
          changed_when: false
          environment:
            CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

        - name: update the placement of grafana hosts
1018
          command: "cephadm shell --fsid {{ fsid }} -- ceph --cluster {{ cluster }} orch apply grafana --placement='{{ groups.get(grafana_server_group_name, []) | length }} label:{{ grafana_server_group_name }}'"
1019
1020
1021
1022
1023
1024
          run_once: true
          changed_when: false
          environment:
            CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

        - name: update the placement of prometheus hosts
1025
          command: "cephadm shell --fsid {{ fsid }} -- ceph --cluster {{ cluster }} orch apply prometheus --placement='{{ groups.get(grafana_server_group_name, []) | length }} label:{{ grafana_server_group_name }}'"
1026
1027
1028
1029
          run_once: true
          changed_when: false
          environment:
            CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039

- name: show ceph orchestrator status
  hosts: "{{ mon_group_name|default('mons') }}"
  become: true
  gather_facts: false
  tasks:
    - import_role:
        name: ceph-defaults

    - name: show ceph orchestrator services
1040
      command: "cephadm shell --fsid {{ fsid }} -- ceph --cluster {{ cluster }} orch ls --refresh"
1041
1042
1043
1044
1045
1046
      run_once: true
      changed_when: false
      environment:
        CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'

    - name: show ceph orchestrator daemons
1047
      command: "cephadm shell --fsid {{ fsid }} -- ceph --cluster {{ cluster }} orch ps --refresh"
1048
1049
1050
1051
      run_once: true
      changed_when: false
      environment:
        CEPHADM_IMAGE: '{{ ceph_docker_registry }}/{{ ceph_docker_image }}:{{ ceph_docker_image_tag }}'
1052
1053
1054
1055
1056
1057
1058

    - name: inform users about cephadm
      debug:
        msg: |
          This Ceph cluster is now managed by cephadm. Any new changes to the
          cluster need to be achieved by using the cephadm CLI and you don't
          need to use ceph-ansible playbooks anymore.