ceph
/
ceph-ansible
mirror of https://github.com/ceph/ceph-ansible.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
ceph-ansible/roles/ceph-handler/tasks/handler_osds.yml

93 lines
3.4 KiB
YAML
Raw Normal View History

ceph-handler: group listen topics and condition We are using multiple listen topics with the handlers. That means that we are notifying 4 tasks for each handler. Instead we can group the listen on an include_tasks and based on the group condition. Before: NOTIFIED HANDLER ceph-handler : set _mon_handler_called before restart for mon0 NOTIFIED HANDLER ceph-handler : copy mon restart script for mon0 NOTIFIED HANDLER ceph-handler : restart ceph mon daemon(s) for mon0 NOTIFIED HANDLER ceph-handler : set _mon_handler_called after restart for mon0 NOTIFIED HANDLER ceph-handler : set _osd_handler_called before restart for mon0 NOTIFIED HANDLER ceph-handler : copy osd restart script for mon0 NOTIFIED HANDLER ceph-handler : restart ceph osds daemon(s) for mon0 NOTIFIED HANDLER ceph-handler : set _osd_handler_called after restart for mon0 NOTIFIED HANDLER ceph-handler : set _mds_handler_called before restart for mon0 NOTIFIED HANDLER ceph-handler : copy mds restart script for mon0 NOTIFIED HANDLER ceph-handler : restart ceph mds daemon(s) for mon0 NOTIFIED HANDLER ceph-handler : set _mds_handler_called after restart for mon0 NOTIFIED HANDLER ceph-handler : set _rgw_handler_called before restart for mon0 NOTIFIED HANDLER ceph-handler : copy rgw restart script for mon0 NOTIFIED HANDLER ceph-handler : restart ceph rgw daemon(s) for mon0 NOTIFIED HANDLER ceph-handler : set _rgw_handler_called after restart for mon0 NOTIFIED HANDLER ceph-handler : set _mgr_handler_called before restart for mon0 NOTIFIED HANDLER ceph-handler : copy mgr restart script for mon0 NOTIFIED HANDLER ceph-handler : restart ceph mgr daemon(s) for mon0 NOTIFIED HANDLER ceph-handler : set _mgr_handler_called after restart for mon0 NOTIFIED HANDLER ceph-handler : set _rbdmirror_handler_called before restart for mon0 NOTIFIED HANDLER ceph-handler : copy rbd mirror restart script for mon0 NOTIFIED HANDLER ceph-handler : restart ceph rbd mirror daemon(s) for mon0 NOTIFIED HANDLER ceph-handler : set _rbdmirror_handler_called after restart for mon0 After: NOTIFIED HANDLER ceph-handler : mons handler for mon0 NOTIFIED HANDLER ceph-handler : osds handler for mon0 NOTIFIED HANDLER ceph-handler : mdss handler for mon0 NOTIFIED HANDLER ceph-handler : rgws handler for mon0 NOTIFIED HANDLER ceph-handler : mgrs handler for mon0 NOTIFIED HANDLER ceph-handler : rbdmirrors handler for mon0 Signed-off-by: Dimitri Savineau <dsavinea@redhat.com> (cherry picked from commit fe9c5b8c686e838a7a16623b5296e9a919fa54de)
2019-10-10 01:58:33 +08:00
---
common: disable/enable pg_autoscaler The PG autoscaler can disrupt the PG checks so the idea here is to disable it and re-enable it back after the restart is done. Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com> (cherry picked from commit 13036115e2862fc8ca9c04e1379fd793e0e7036a)
2021-06-15 00:01:41 +08:00
- name: set_fact trigger_restart
ceph-handler: group listen topics and condition We are using multiple listen topics with the handlers. That means that we are notifying 4 tasks for each handler. Instead we can group the listen on an include_tasks and based on the group condition. Before: NOTIFIED HANDLER ceph-handler : set _mon_handler_called before restart for mon0 NOTIFIED HANDLER ceph-handler : copy mon restart script for mon0 NOTIFIED HANDLER ceph-handler : restart ceph mon daemon(s) for mon0 NOTIFIED HANDLER ceph-handler : set _mon_handler_called after restart for mon0 NOTIFIED HANDLER ceph-handler : set _osd_handler_called before restart for mon0 NOTIFIED HANDLER ceph-handler : copy osd restart script for mon0 NOTIFIED HANDLER ceph-handler : restart ceph osds daemon(s) for mon0 NOTIFIED HANDLER ceph-handler : set _osd_handler_called after restart for mon0 NOTIFIED HANDLER ceph-handler : set _mds_handler_called before restart for mon0 NOTIFIED HANDLER ceph-handler : copy mds restart script for mon0 NOTIFIED HANDLER ceph-handler : restart ceph mds daemon(s) for mon0 NOTIFIED HANDLER ceph-handler : set _mds_handler_called after restart for mon0 NOTIFIED HANDLER ceph-handler : set _rgw_handler_called before restart for mon0 NOTIFIED HANDLER ceph-handler : copy rgw restart script for mon0 NOTIFIED HANDLER ceph-handler : restart ceph rgw daemon(s) for mon0 NOTIFIED HANDLER ceph-handler : set _rgw_handler_called after restart for mon0 NOTIFIED HANDLER ceph-handler : set _mgr_handler_called before restart for mon0 NOTIFIED HANDLER ceph-handler : copy mgr restart script for mon0 NOTIFIED HANDLER ceph-handler : restart ceph mgr daemon(s) for mon0 NOTIFIED HANDLER ceph-handler : set _mgr_handler_called after restart for mon0 NOTIFIED HANDLER ceph-handler : set _rbdmirror_handler_called before restart for mon0 NOTIFIED HANDLER ceph-handler : copy rbd mirror restart script for mon0 NOTIFIED HANDLER ceph-handler : restart ceph rbd mirror daemon(s) for mon0 NOTIFIED HANDLER ceph-handler : set _rbdmirror_handler_called after restart for mon0 After: NOTIFIED HANDLER ceph-handler : mons handler for mon0 NOTIFIED HANDLER ceph-handler : osds handler for mon0 NOTIFIED HANDLER ceph-handler : mdss handler for mon0 NOTIFIED HANDLER ceph-handler : rgws handler for mon0 NOTIFIED HANDLER ceph-handler : mgrs handler for mon0 NOTIFIED HANDLER ceph-handler : rbdmirrors handler for mon0 Signed-off-by: Dimitri Savineau <dsavinea@redhat.com> (cherry picked from commit fe9c5b8c686e838a7a16623b5296e9a919fa54de)
2019-10-10 01:58:33 +08:00
set_fact:
common: disable/enable pg_autoscaler The PG autoscaler can disrupt the PG checks so the idea here is to disable it and re-enable it back after the restart is done. Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com> (cherry picked from commit 13036115e2862fc8ca9c04e1379fd793e0e7036a)
2021-06-15 00:01:41 +08:00
trigger_restart: true
loop: "{{ groups[osd_group_name] }}"
when: hostvars[item]['handler_osd_status'] | default(False) | bool
handler: fix call to container_exec_cmd in handler_osds When unsetting the noup flag, we must call container_exec_cmd from the delegated node (first mon member) Also, adding a `run_once: true` because this task needs to be run only 1 time. Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1792320 Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com> (cherry picked from commit 22865cde9c7cb3be30c359ce8679ec841b16a663)
2020-01-17 22:50:40 +08:00
run_once: true
osd: support scaling up using --limit This commit lets add-osd.yml in place but mark the deprecation of the playbook. Scaling up OSDs is now possible using --limit Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com> (cherry picked from commit 3496a0efa2d7ae2872476387e9d801fb32414f63)
2019-07-09 21:40:47 +08:00
common: disable/enable pg_autoscaler The PG autoscaler can disrupt the PG checks so the idea here is to disable it and re-enable it back after the restart is done. Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com> (cherry picked from commit 13036115e2862fc8ca9c04e1379fd793e0e7036a)
2021-06-15 00:01:41 +08:00
- name: osd handler
when: trigger_restart | default(False) | bool
block:
- name: get pool list
osds: use osd pool ls instead of osd dump command The ceph osd pool ls detail command is a subset of the ceph osd dump command. $ ceph osd dump --format json|wc -c 10117 $ ceph osd pool ls detail --format json|wc -c 4740 Signed-off-by: Dimitri Savineau <dsavinea@redhat.com> (cherry picked from commit 06471a4b82d63ebb35f80d45aa6ae629a4daeedc)
2021-07-29 02:54:15 +08:00
command: "{{ ceph_cmd }} --cluster {{ cluster }} osd pool ls detail -f json"
common: disable/enable pg_autoscaler The PG autoscaler can disrupt the PG checks so the idea here is to disable it and re-enable it back after the restart is done. Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com> (cherry picked from commit 13036115e2862fc8ca9c04e1379fd793e0e7036a)
2021-06-15 00:01:41 +08:00
register: pool_list
delegate_to: "{{ groups.get(mon_group_name, [])[0] }}"
run_once: true
changed_when: false
ceph-handler: Fix osd handler in check mode Run the Ceph commands that only gather information (without making any changes to the cluster) when running Ansible in check mode. This allows the tasks that depend on the variables set by those tasks to succeed in check mode. Signed-off-by: Benoît Knecht <bknecht@protonmail.ch> (cherry picked from commit 498acd7527410f7f359b2b0181e83ca39c682ec0)
2021-07-26 19:03:56 +08:00
check_mode: false
common: disable/enable pg_autoscaler The PG autoscaler can disrupt the PG checks so the idea here is to disable it and re-enable it back after the restart is done. Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com> (cherry picked from commit 13036115e2862fc8ca9c04e1379fd793e0e7036a)
2021-06-15 00:01:41 +08:00
- name: get balancer module status
command: "{{ ceph_cmd }} --cluster {{ cluster }} balancer status -f json"
register: balancer_status
run_once: true
delegate_to: "{{ groups[mon_group_name][0] }}"
changed_when: false
ceph-handler: Fix osd handler in check mode Run the Ceph commands that only gather information (without making any changes to the cluster) when running Ansible in check mode. This allows the tasks that depend on the variables set by those tasks to succeed in check mode. Signed-off-by: Benoît Knecht <bknecht@protonmail.ch> (cherry picked from commit 498acd7527410f7f359b2b0181e83ca39c682ec0)
2021-07-26 19:03:56 +08:00
check_mode: false
common: disable/enable pg_autoscaler The PG autoscaler can disrupt the PG checks so the idea here is to disable it and re-enable it back after the restart is done. Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com> (cherry picked from commit 13036115e2862fc8ca9c04e1379fd793e0e7036a)
2021-06-15 00:01:41 +08:00
- name: set_fact pools_pgautoscaler_mode
set_fact:
pools_pgautoscaler_mode: "{{ pools_pgautoscaler_mode | default([]) | union([{'name': item.pool_name, 'mode': item.pg_autoscale_mode}]) }}"
run_once: true
osds: use osd pool ls instead of osd dump command The ceph osd pool ls detail command is a subset of the ceph osd dump command. $ ceph osd dump --format json|wc -c 10117 $ ceph osd pool ls detail --format json|wc -c 4740 Signed-off-by: Dimitri Savineau <dsavinea@redhat.com> (cherry picked from commit 06471a4b82d63ebb35f80d45aa6ae629a4daeedc)
2021-07-29 02:54:15 +08:00
with_items: "{{ pool_list.stdout | default('{}') | from_json }}"
common: disable/enable pg_autoscaler The PG autoscaler can disrupt the PG checks so the idea here is to disable it and re-enable it back after the restart is done. Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com> (cherry picked from commit 13036115e2862fc8ca9c04e1379fd793e0e7036a)
2021-06-15 00:01:41 +08:00
- name: disable balancer
command: "{{ ceph_cmd }} --cluster {{ cluster }} balancer off"
run_once: true
delegate_to: "{{ groups[mon_group_name][0] }}"
changed_when: false
when: (balancer_status.stdout | from_json)['active'] | bool
- name: disable pg autoscale on pools
command: "{{ ceph_cmd }} --cluster {{ cluster }} osd pool set {{ item.name }} pg_autoscale_mode off"
with_items: "{{ pools_pgautoscaler_mode }}"
delegate_to: "{{ groups.get(mon_group_name, [])[0] }}"
run_once: true
when:
- pools_pgautoscaler_mode is defined
- item.mode == 'on'
ceph-handler: group listen topics and condition We are using multiple listen topics with the handlers. That means that we are notifying 4 tasks for each handler. Instead we can group the listen on an include_tasks and based on the group condition. Before: NOTIFIED HANDLER ceph-handler : set _mon_handler_called before restart for mon0 NOTIFIED HANDLER ceph-handler : copy mon restart script for mon0 NOTIFIED HANDLER ceph-handler : restart ceph mon daemon(s) for mon0 NOTIFIED HANDLER ceph-handler : set _mon_handler_called after restart for mon0 NOTIFIED HANDLER ceph-handler : set _osd_handler_called before restart for mon0 NOTIFIED HANDLER ceph-handler : copy osd restart script for mon0 NOTIFIED HANDLER ceph-handler : restart ceph osds daemon(s) for mon0 NOTIFIED HANDLER ceph-handler : set _osd_handler_called after restart for mon0 NOTIFIED HANDLER ceph-handler : set _mds_handler_called before restart for mon0 NOTIFIED HANDLER ceph-handler : copy mds restart script for mon0 NOTIFIED HANDLER ceph-handler : restart ceph mds daemon(s) for mon0 NOTIFIED HANDLER ceph-handler : set _mds_handler_called after restart for mon0 NOTIFIED HANDLER ceph-handler : set _rgw_handler_called before restart for mon0 NOTIFIED HANDLER ceph-handler : copy rgw restart script for mon0 NOTIFIED HANDLER ceph-handler : restart ceph rgw daemon(s) for mon0 NOTIFIED HANDLER ceph-handler : set _rgw_handler_called after restart for mon0 NOTIFIED HANDLER ceph-handler : set _mgr_handler_called before restart for mon0 NOTIFIED HANDLER ceph-handler : copy mgr restart script for mon0 NOTIFIED HANDLER ceph-handler : restart ceph mgr daemon(s) for mon0 NOTIFIED HANDLER ceph-handler : set _mgr_handler_called after restart for mon0 NOTIFIED HANDLER ceph-handler : set _rbdmirror_handler_called before restart for mon0 NOTIFIED HANDLER ceph-handler : copy rbd mirror restart script for mon0 NOTIFIED HANDLER ceph-handler : restart ceph rbd mirror daemon(s) for mon0 NOTIFIED HANDLER ceph-handler : set _rbdmirror_handler_called after restart for mon0 After: NOTIFIED HANDLER ceph-handler : mons handler for mon0 NOTIFIED HANDLER ceph-handler : osds handler for mon0 NOTIFIED HANDLER ceph-handler : mdss handler for mon0 NOTIFIED HANDLER ceph-handler : rgws handler for mon0 NOTIFIED HANDLER ceph-handler : mgrs handler for mon0 NOTIFIED HANDLER ceph-handler : rbdmirrors handler for mon0 Signed-off-by: Dimitri Savineau <dsavinea@redhat.com> (cherry picked from commit fe9c5b8c686e838a7a16623b5296e9a919fa54de)
2019-10-10 01:58:33 +08:00
# This does not just restart OSDs but everything else too. Unfortunately
common: disable/enable pg_autoscaler The PG autoscaler can disrupt the PG checks so the idea here is to disable it and re-enable it back after the restart is done. Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com> (cherry picked from commit 13036115e2862fc8ca9c04e1379fd793e0e7036a)
2021-06-15 00:01:41 +08:00
# at this time the ansible role does not have an OSD id list to use
# for restarting them specifically.
# This does not need to run during a rolling update as the playbook will
# restart all OSDs using the tasks "start ceph osd" or
# "restart containerized ceph osd"
- name: copy osd restart script
template:
src: restart_osd_daemon.sh.j2
dest: /tmp/restart_osd_daemon.sh
owner: root
group: root
mode: 0750
- name: restart ceph osds daemon(s)
command: /usr/bin/env bash /tmp/restart_osd_daemon.sh
when:
- hostvars[item]['handler_osd_status'] | default(False) | bool
- handler_health_osd_check | bool
- hostvars[item]['_osd_handler_called'] | default(False) | bool
with_items: "{{ groups[osd_group_name] | intersect(ansible_play_batch) }}"
delegate_to: "{{ item }}"
run_once: True
- name: set _osd_handler_called after restart
set_fact:
_osd_handler_called: False
- name: re-enable pg autoscale on pools
command: "{{ ceph_cmd }} --cluster {{ cluster }} osd pool set {{ item.name }} pg_autoscale_mode on"
with_items: "{{ pools_pgautoscaler_mode }}"
delegate_to: "{{ groups.get(mon_group_name, [])[0] }}"
run_once: true
when:
- pools_pgautoscaler_mode is defined
- item.mode == 'on'
- name: re-enable balancer
command: "{{ ceph_cmd }} --cluster {{ cluster }} balancer on"
run_once: true
delegate_to: "{{ groups[mon_group_name][0] }}"
changed_when: false
when: (balancer_status.stdout | from_json)['active'] | bool