osd: ensure /var/lib/ceph/osd/{cluster}-{id} is present

This commit ensures that the `/var/lib/ceph/osd/{{ cluster }}-{{ osd_id }}` is
present before starting OSDs.

This is needed specificly when redeploying an OSD in case of OS upgrade
failure.
Since ceph data are still present on its devices then the node can be
redeployed, however those directories aren't present since they are
initially created by ceph-volume. We could recreate them manually but
for better user experience we can ask ceph-ansible to recreate them.

NOTE:
this only works for OSDs that were deployed with ceph-volume.
ceph-disk deployed OSDs would have to get those directories recreated
manually.

Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1898486
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit 873fc8ec0ff12fa1d1b45c5400050f15d0417480)

diff --git a/roles/ceph-osd/tasks/start_osds.yml b/roles/ceph-osd/tasks/start_osds.yml
index b073ac1bab78c8f252a1fbf0abf7f99242d98aaf..5397d5841330c0df02b99375c9fb7b7f6b479087 100644 (file)
--- a/roles/ceph-osd/tasks/start_osds.yml
+++ b/roles/ceph-osd/tasks/start_osds.yml
@@ -51,6 +51,15 @@
     - ceph_osd_systemd_overrides is defined
     - ansible_service_mgr == 'systemd'
 
+- name: ensure "/var/lib/ceph/osd/{{ cluster }}-{{ item }}" is present
+  file:
+    state: directory
+    path: "/var/lib/ceph/osd/{{ cluster }}-{{ item }}"
+    mode: "{{ ceph_directories_mode }}"
+    owner: "{{ ceph_uid if containerized_deployment | bool else 'ceph' }}"
+    group: "{{ ceph_uid if containerized_deployment | bool else 'ceph' }}"
+  with_items: "{{ ((ceph_osd_ids.stdout | default('{}') | from_json).keys() | list) | union(osd_ids_non_container.stdout_lines | default([])) }}"
+
 - name: systemd start osd
   systemd:
     name: ceph-osd@{{ item }}