Revert "Failed_when and stat module"

cluster-os-migration.yml

@@ -21,7 +21,7 @@
       stat: >
         path=/var/lib/ceph/mon/ceph-{{ ansible_hostname }}/migration_completed
       register: migration_completed
-      failed_when: false
+      ignore_errors: True
 
     - name: Check for failed run
       stat: >
@@ -248,7 +248,7 @@
       stat: >
         path=/var/lib/ceph/migration_completed
       register: migration_completed
-      failed_when: false
+      ignore_errors: True
 
     - name: Check for failed run
       stat: >
@@ -286,13 +286,13 @@
     - name: Check if sysvinit
       shell: stat /var/lib/ceph/osd/ceph-*/sysvinit
       register: osdsysvinit
-      failed_when: false
+      ignore_errors: True
       changed_when: False
 
     - name: Check if upstart
       shell: stat /var/lib/ceph/osd/ceph-*/upstart
       register: osdupstart
-      failed_when: false
+      ignore_errors: True
       changed_when: False
 
     - name: Archive ceph configs
@@ -443,7 +443,7 @@
       stat: >
         path=/var/lib/ceph/radosgw/migration_completed
       register: migration_completed
-      failed_when: false
+      ignore_errors: True
 
     - name: Check for failed run
       stat: >

contrib/splitup.yml

@@ -50,7 +50,7 @@
       tags: update
       command: >
         git remote add upstream git@github.com:ceph/ceph-ansible.git
-      failed_when: false
+      ignore_errors: yes
 
     - name: pulles the newest changes from upstream
       tags: update

purge-cluster.yml

@@ -15,12 +15,12 @@
   - name: disk zap
     command: /usr/sbin/sgdisk --zap-all --clear --mbrtogpt -g -- {{ item }}
     with_items: devices
-    failed_when: false
+    ignore_errors: true
 
   - name: disk zap
     command: /usr/sbin/sgdisk --zap-all --clear --mbrtogpt -g -- {{ item }}
     with_items: devices
-    failed_when: false
+    ignore_errors: true
 
   - name: call partprobe
     command: partprobe
@@ -31,7 +31,7 @@
 
   - name: remove osd data
     shell: rm -rf /var/lib/ceph/osd/*/*
-    failed_when: false
+    ignore_errors: true
 
   - name: purge remaining data
     command: ceph-deploy purgedata {{ ansible_fqdn }}

roles/ceph-common/tasks/main.yml

@@ -26,15 +26,15 @@
   when: ansible_os_family == 'Debian'
 
 - name: check for a ceph socket
-  stat: path=/var/run/ceph/*.asok
+  shell: "stat /var/run/ceph/*.asok > /dev/null 2>&1"
   changed_when: false
-  failed_when: false
+  ignore_errors: true
   register: socket
 
 - name: check for a rados gateway socket
-  stat: path={{ rbd_client_admin_socket_path }}*.asok
+  shell: "stat {{ rbd_client_admin_socket_path }}*.asok > /dev/null 2>&1"
   changed_when: false
-  failed_when: false
+  ignore_errors: true
   register: socketrgw
 
 - name: create a local fetch directory if it doesn't exist

roles/ceph-common/tasks/misc/system_tuning.yml

@@ -2,18 +2,18 @@
 - name: disable osd directory parsing by updatedb
   command: updatedb -e /var/lib/ceph
   changed_when: false
-  failed_when: false
+  ignore_errors: true
 
 - name: disable transparent hugepage
   command: "echo never > /sys/kernel/mm/transparent_hugepage/enabled"
   changed_when: false
-  failed_when: false
+  ignore_errors: true
   when: disable_transparent_hugepage
 
 - name: disable swap
   command: swapoff -a
   changed_when: false
-  failed_when: false
+  ignore_errors: true
   when: disable_swap
 
 - name: apply operating system tuning

roles/ceph-common/tasks/pre_requisites/prerequisite_rh_storage_cdn_install.yml

@@ -7,7 +7,7 @@
 - name: check if the red hat storage monitor repo is already present
   shell: yum --noplugins --cacheonly repolist | grep -sq rhel-7-server-rhceph-1.3-mon-rpms
   changed_when: false
-  failed_when: false
+  ignore_errors: true
   register: rh_storage_mon_repo
   when: mon_group_name in group_names
 
@@ -21,7 +21,7 @@
 - name: check if the red hat storage osd repo is already present
   shell: yum --noplugins --cacheonly repolist | grep -sq rhel-7-server-rhceph-1.3-osd-rpms
   changed_when: false
-  failed_when: false
+  ignore_errors: true
   register: rh_storage_osd_repo
   when: osd_group_name in group_names
 
@@ -35,7 +35,7 @@
 - name: check if the red hat storage rados gateway repo is already present
   shell: yum --noplugins --cacheonly repolist | grep -sq rhel-7-server-rhceph-1.3-tools-rpms
   changed_when: false
-  failed_when: false
+  ignore_errors: true
   register: rh_storage_rgw_repo
   when: rgw_group_name in group_names
 

roles/ceph-mds/tasks/docker/fetch_configs.yml

@@ -13,7 +13,7 @@
   with_items: ceph_config_keys
   changed_when: false
   sudo: false
-  failed_when: false
+  ignore_errors: true
   register: statconfig
 
 - name: try to fetch ceph config and keys

roles/ceph-mon/tasks/docker.yml

@@ -30,7 +30,7 @@
   with_items: ceph_config_keys
   changed_when: false
   sudo: false
-  failed_when: false
+  ignore_errors: true
   register: statconfig
 
 - name: try to fetch ceph config and keys

roles/ceph-mon/tasks/docker/fetch_configs.yml

@@ -15,7 +15,7 @@
   with_items: ceph_config_keys
   changed_when: false
   sudo: false
-  failed_when: false
+  ignore_errors: true
   register: statconfig
 
 - name: try to fetch ceph config and keys

roles/ceph-mon/tasks/openstack_config.yml

@@ -7,7 +7,7 @@
     - "{{ openstack_nova_pool }}"
     - "{{ openstack_cinder_backup_pool }}"
   changed_when: false
-  failed_when: false
+  ignore_errors: true
 
 - name: create openstack keys
   command: >

roles/ceph-osd/tasks/activate_osds.yml

@@ -9,7 +9,7 @@
     - ispartition.results
     - devices
   changed_when: false
-  failed_when: false 
+  ignore_errors: true
   when:
     item.0.rc == 0 and
     item.1.rc != 0
@@ -21,7 +21,7 @@
     - ispartition.results
     - devices
   changed_when: false
-  failed_when: false
+  ignore_errors: true
   when: item.0.rc == 0
 
 - include: osd_fragment.yml

roles/ceph-osd/tasks/check_devices.yml

@@ -2,18 +2,18 @@
 # NOTE (leseb): current behavior of ceph-disk is to fail when the device is mounted "stderr: ceph-disk: Error: Device is mounted: /dev/sdb1"
 # the return code is 1, which makes sense, however ideally if ceph-disk will detect a ceph partition
 # it should exist we rc=0 and don't do anything unless we do something like --force
-# As as a final word, I prefer to keep the partition check instead of running ceph-disk prepare with "failed_when: false"
+# As as a final word, I prefer to keep the partition check instead of running ceph-disk prepare with "ignore_errors: true"
 # I believe it's safer
 - name: check if the device is a partition or a disk
   shell: "echo '{{ item }}' | egrep '/dev/(([a-z]{3,4}[0-9]$)|(cciss/c[0-9]{1}d[0-9]{1}p[0-9]$))'"
   with_items: devices
   changed_when: false
-  failed_when: false
+  ignore_errors: true
   register: ispartition
 
 - name: if partition named 'ceph' exists
   shell: "parted --script {{ item }} print | egrep -sq '^ 1.*ceph'"
   with_items: devices
   changed_when: false
-  failed_when: false
+  ignore_errors: true
   register: parted

roles/ceph-osd/tasks/docker/fetch_configs.yml

@@ -10,7 +10,7 @@
   with_items: ceph_config_keys
   changed_when: false
   sudo: false
-  failed_when: false
+  ignore_errors: true
   register: statconfig
 
 - name: try to fetch ceph config and keys

roles/ceph-osd/tasks/osd_fragment.yml

@@ -3,14 +3,14 @@
   shell: "df | grep {{ item }} | awk '{print $6}'"
   with_items: devices
   changed_when: false
-  failed_when: false
+  ignore_errors: true
   register: osd_path
 
 - name: get osd id
   command: cat {{ item.stdout }}/whoami
   with_items: osd_path.results
   changed_when: false
-  failed_when: false
+  ignore_errors: true
   register: osd_id
 
 - name: create a ceph fragment and assemble directory

roles/ceph-osd/tasks/scenarios/docker.yml

@@ -18,7 +18,7 @@
   stat: >
     path={{ item }}
   with_items: ceph_config_keys
-  failed_when: false
+  ignore_errors: true
   register: statconfig
 
 - name: try to fetch ceph config and keys

roles/ceph-restapi/tasks/start_restapi.yml

@@ -2,7 +2,7 @@
 - name: check if ceph rest api is already started
   shell: "pgrep ceph-rest-api"
   changed_when: false
-  failed_when: false
+  ignore_errors: true
   register: restapi_status
 
 - name: start ceph rest api

roles/ceph-rgw/tasks/installs/install_debian.yml

@@ -131,7 +131,7 @@
     - a2ensite rgw.conf
     - a2dissite *default
   changed_when: false
-  failed_when: false
+  ignore_errors: true
   notify:
     - restart apache2
 

roles/ceph-rgw/tasks/start_radosgw.yml

@@ -2,7 +2,7 @@
 - name: check if rgw is started
   command: /etc/init.d/radosgw status
   register: rgwstatus
-  failed_when: false
+  ignore_errors: true
 
 - name: start rgw
   command: /etc/init.d/radosgw start

rolling_update.yml

@@ -92,12 +92,12 @@
     - name: Check if sysvinit
       shell: stat /var/lib/ceph/osd/ceph-*/sysvinit
       register: osdsysvinit
-      failed_when: false
+      ignore_errors: True
 
     - name: Check if upstart
       shell: stat /var/lib/ceph/osd/ceph-*/upstart
       register: osdupstart
-      failed_when: false
+      ignore_errors: True
 
     - name: Gracefully stop the OSDs (Upstart)
       service: >