DEv-452: added recursive _set_server_state.yml to work around hetzners...

galaxy-requirements.yml

@@ -24,12 +24,8 @@ roles:
   version: 3.3.2
 
 collections:
-#- name: hetzner.hcloud
-#  version: 1.6.0
 - name: hetzner.hcloud
-  src: https://github.com/fredFromJupiter/hetzner.hcloud.git
-  github_branch: "fredFromJupiter-patch-1"
-  scm: git
+  version: 1.6.0
 - name: community.general
 - name: community.docker
   version: 2.1.1

restore-remote-database-backup.yml

@@ -68,6 +68,8 @@
 
   roles:
     - role: hcloud
+      vars:
+        sma_digitalocean_ttl: 60 # set it to 60sec to reduce DNS caching problems with internal IT in case of debugging ansible problems ;)
 
 #############################################################
 #   Provisioning server(s) for created inventory

roles/backup/files/push_backups_to_restore_server.sh

@@ -17,6 +17,9 @@ BACKUP_FILE_FOR_TRANSFER=$(find "${LOCAL_BACKUP_DIR}/${DATE}/" -name *.gz.gpg |
 REMOTE_BACKUP_DIR="/home/${REMOTE_SYSTEM_USER}/backups/${STAGE}/${DATABASE_ENGINE}"
 DEST_DIR="${REMOTE_BACKUP_DIR}/${DATE}/"
 
+# avoid "REMOTE HOST IDENTIFICATION HAS CHANGED" - errors due to dynamic created server on restore process
+ssh-keygen -f "/home/backuphamster/.ssh/known_hosts" -R ${DATABASE_SERVER_IP}
+
 SSH_OPTIONS='-o StrictHostKeyChecking=no'
 
 # needed due to unknown rsync option --mkpath in rsync version 3.1.3

roles/hcloud/tasks/_set_server_state.yml

@@ -1,4 +1,12 @@
 ---
+- name: "Block to handle hetzner server state in case of problems"
+  vars:
+    max_retries: "{{ max_retries | default(15) | int }}"
+    retry_delay: "{{ retry_delay | default(60) | int }}"
+  block:
+  - name: "Increment the retry count"
+    set_fact:
+      retry_count: "{{ retry_count | default(0) | int + 1 }}"
 
   - name: "Checking state for server <{{ inventory_hostname }}> is <{{ server_state }}>"
     hetzner.hcloud.hcloud_server:
@@ -17,7 +25,12 @@
     register: hcloud_response
     ignore_errors: yes
 
-# poweron server if server not running
+  - name: "Block - DEBUG: hcloud_response"
+    debug:
+      msg: '{{ hcloud_response.msg }}'
+    when:
+      - hcloud_response.msg is defined
+
   - name: "Ensure Server is STARTED when server_state=present"
     hetzner.hcloud.hcloud_server:
       api_token: "{{ hetzner_authentication_ansible }}"
@@ -25,45 +38,27 @@
       state: "started"
     delegate_to: 127.0.0.1
     become: false
-  async: 300
+    async: 150
     poll: 15
+    register: hcloud_response
     when:
       - server_state == 'present'
-    - not hcloud_response.msg is defined
 
-# just catching hetzner error and trying to avoid manual interaction
-- name: "DEBUGGING Block for hetzner << message not found >> - error"
-  block:
-  - name: "DEBUG: hcloud_response"
+  rescue:
+    - name: "RESCUE - fail: Maximum retries reached"
+      fail:
+        msg: "max_retries of {{ max_retries }} reached. Plz check."
+      when: retry_count | int == max_retries | int
+
+    - name: "RESCUE-fail DEBUG: hcloud_response"
       debug:
         msg: '{{ hcloud_response.msg }}'
 
-  - name: "Ensure Server is STARTED when server_state=present"
-    hetzner.hcloud.hcloud_server:
-      api_token: "{{ hetzner_authentication_ansible }}"
-      name: "{{ inventory_hostname }}"
-      state: "started"
-    delegate_to: 127.0.0.1
-    become: false
-    retries: 40
-    delay: 15
-    until: server_locked.rc == 0
-    register: server_locked
-    when:
-      - server_state == 'present'
-      - hcloud_response.msg == 'cannot perform operation because server is locked'
-
-  - name: "Ensure Server is STARTED when server_state=present"
-    hetzner.hcloud.hcloud_server:
-      api_token: "{{ hetzner_authentication_ansible }}"
-      name: "{{ inventory_hostname }}"
-      state: "started"
-    delegate_to: 127.0.0.1
+    - name: "RESCUE: wait_for between retries"
+      wait_for:
+        timeout: "{{ retry_delay }}"
+      delegate_to: localhost
       become: false
-    async: 300
-    poll: 15
-    when:
-      - server_state == 'present'
 
-  when:
-    - hcloud_response.msg is defined
+    - name: "Include _set_server one time again - retry_count: {{ retry_count }}"
+      include_tasks: _set_server_state.yml

roles/sma_digitalocean/tasks/domain.yml

@@ -68,7 +68,7 @@
   vars:
     create_record:
         type: 'A'
-        ttl: 1800
+        ttl: "{{ sma_digitalocean_ttl | default(1800) }}"
         data: "{{ record_data }}"
         name: "{{ record_name }}"
   when:

smardigo/backup/process/primary-process.bpmn

@@ -80,7 +80,7 @@
     <bpmn2:startEvent id="Event_0wqvypp">
       <bpmn2:outgoing>Flow_1a8vxf0</bpmn2:outgoing>
       <bpmn2:timerEventDefinition id="TimerEventDefinition_1geoi5v">
-        <bpmn2:timeCycle xsi:type="bpmn2:tFormalExpression">0 0 0 * * ?</bpmn2:timeCycle>
+        <bpmn2:timeCycle xsi:type="bpmn2:tFormalExpression">0 0 * * * ?</bpmn2:timeCycle>
       </bpmn2:timerEventDefinition>
     </bpmn2:startEvent>
     <bpmn2:sequenceFlow id="Flow_0zzip76" sourceRef="Gateway_1f7y3mt" targetRef="Activity_12c4494" />