osds: use pg stat command instead of ceph status

The ceph status command returns a lot of information stored in variables
and/or facts which could consume resources for nothing.
When checking the pgs state, we're using the pgmap structure in the ceph
status output.
To optimize this, we could use the ceph pg stat command which contains
the same needed information.
This command returns less information (only about pgs) and is slightly
faster than the ceph status command.

$ ceph status -f json | wc -c
2000
$ ceph pg stat -f json | wc -c
240
$ time ceph status -f json > /dev/null

real	0m0.529s
user	0m0.503s
sys	0m0.024s
$ time ceph pg stat -f json > /dev/null

real	0m0.426s
user	0m0.409s
sys	0m0.016s

The data returned by the ceph status is even bigger when using the
nautilus release.

$ ceph status -f json | wc -c
35005
$ ceph pg stat -f json | wc -c
240

Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit ee505885908ac2ae15bf201a638359faaf78d251)

diff --git a/infrastructure-playbooks/cephadm-adopt.yml b/infrastructure-playbooks/cephadm-adopt.yml
index 14ced297c5058050e44b514ca11961b5b87248d6..fc5be11a2f21613f884c38926a849521e0efcf56 100644 (file)
--- a/infrastructure-playbooks/cephadm-adopt.yml
+++ b/infrastructure-playbooks/cephadm-adopt.yml
@@ -479,13 +479,13 @@
       when: not containerized_deployment | bool
 
     - name: waiting for clean pgs...
-      command: "{{ cephadm_cmd }} shell --fsid {{ fsid }} -- ceph --cluster {{ cluster }} -s --format json"
+      command: "{{ cephadm_cmd }} shell --fsid {{ fsid }} -- ceph --cluster {{ cluster }} pg stat --format json"
       changed_when: false
       register: ceph_health_post
       until: >
-        (((ceph_health_post.stdout | from_json).pgmap.pgs_by_state | length) > 0)
+        (((ceph_health_post.stdout | from_json).pg_summary.num_pg_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)
+        (((ceph_health_post.stdout | from_json).pg_summary.num_pg_by_state | selectattr('name', 'search', '^active\\+clean') | map(attribute='num') | list | sum) == (ceph_health_post.stdout | from_json).pg_summary.num_pgs)
       delegate_to: "{{ groups[mon_group_name][0] }}"
       retries: "{{ health_osd_check_retries }}"
       delay: "{{ health_osd_check_delay }}"

diff --git a/infrastructure-playbooks/rolling_update.yml b/infrastructure-playbooks/rolling_update.yml
index 6a2f1fbc3a48064ba3a4591fa1e0b21b7eeff033..4c46b6369a851f907a66838388bf127e201b0a1c 100644 (file)
--- a/infrastructure-playbooks/rolling_update.yml
+++ b/infrastructure-playbooks/rolling_update.yml
@@ -415,21 +415,21 @@
         - not containerized_deployment | bool
 
     - name: get num_pgs - non container
-      command: "{{ container_exec_cmd_update_osd|default('') }} ceph --cluster {{ cluster }} -s --format json"
+      command: "{{ container_exec_cmd_update_osd|default('') }} ceph --cluster {{ cluster }} pg stat --format json"
       register: ceph_pgs
       delegate_to: "{{ groups[mon_group_name][0] }}"
 
     - name: waiting for clean pgs...
-      command: "{{ container_exec_cmd_update_osd|default('') }} ceph --cluster {{ cluster }} -s --format json"
+      command: "{{ container_exec_cmd_update_osd|default('') }} ceph --cluster {{ cluster }} pg stat --format json"
       register: ceph_health_post
       until: >
-        (((ceph_health_post.stdout | from_json).pgmap.pgs_by_state | length) > 0)
+        (((ceph_health_post.stdout | from_json).pg_summary.num_pg_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_pgs.stdout | from_json).pgmap.num_pgs)
+        (((ceph_health_post.stdout | from_json).pg_summary.num_pg_by_state | selectattr('name', 'search', '^active\\+clean') | map(attribute='num') | list | sum) == (ceph_pgs.stdout | from_json).pg_summary.num_pgs)
       delegate_to: "{{ groups[mon_group_name][0] }}"
       retries: "{{ health_osd_check_retries }}"
       delay: "{{ health_osd_check_delay }}"
-      when: (ceph_pgs.stdout | from_json).pgmap.num_pgs != 0
+      when: (ceph_pgs.stdout | from_json).pg_summary.num_pgs != 0
 
 
 - name: complete osd upgrade

diff --git a/infrastructure-playbooks/switch-from-non-containerized-to-containerized-ceph-daemons.yml b/infrastructure-playbooks/switch-from-non-containerized-to-containerized-ceph-daemons.yml
index 4f4e6d531b4ef2ce7a2656b66306e73dee7ffd56..eab634bd97bd99d12452bfa0f7c8c135c8bd83bd 100644 (file)
--- a/infrastructure-playbooks/switch-from-non-containerized-to-containerized-ceph-daemons.yml
+++ b/infrastructure-playbooks/switch-from-non-containerized-to-containerized-ceph-daemons.yml
@@ -320,12 +320,12 @@
   post_tasks:
     - name: container - waiting for clean pgs...
       command: >
-        {{ container_binary }} exec ceph-mon-{{ hostvars[groups[mon_group_name][0]]['ansible_hostname'] }} ceph --cluster {{ cluster }} -s --format json
+        {{ container_binary }} exec ceph-mon-{{ hostvars[groups[mon_group_name][0]]['ansible_hostname'] }} ceph --cluster {{ cluster }} pg stat --format json
       register: ceph_health_post
       until: >
-        (((ceph_health_post.stdout | from_json).pgmap.pgs_by_state | length) > 0)
+        (((ceph_health_post.stdout | from_json).pg_summary.num_pg_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)
+        (((ceph_health_post.stdout | from_json).pg_summary.num_pg_by_state | selectattr('name', 'search', '^active\\+clean') | map(attribute='num') | list | sum) == (ceph_health_post.stdout | from_json).pg_summary.num_pgs)
       delegate_to: "{{ groups[mon_group_name][0] }}"
       retries: "{{ health_osd_check_retries }}"
       delay: "{{ health_osd_check_delay }}"