group_vars/all

@@ -58,7 +58,7 @@ system_maintenance_timeout_cleanup_services:  "15min"
 system_maintenance_timeout_backup_services:   "1h"
 system_maintenance_timeout_heal_docker:       "30min"
 system_maintenance_timeout_update_docker:     "5min"
-system_maintenance_timeout_defroster:         "2min"
+system_maintenance_timeout_defroster:    "10min"
 
 ## Services
 

roles/backup-docker-to-local/tasks/main.yml

@@ -30,7 +30,7 @@
 - name: configure backup-docker-to-local.timer.tpl
   template: src=backup-docker-to-local.timer.j2 dest=/etc/systemd/system/backup-docker-to-local.timer
   register: backup_docker_to_local_timer
-  changed_when: backup_docker_to_local_timer.changed or activate_all_timers | bool
+  changed_when: backup_docker_to_local_timer.changed or activate_all_timers | default(false) | bool
   notify: restart backup-docker-to-local.timer
   when: run_once_backup_docker_to_local is not defined
 

roles/backup-remote-to-local/tasks/main.yml

@@ -17,7 +17,7 @@
 - name: create backup-remote-to-local.timer
   template: src=backup-remote-to-local.timer.j2 dest=/etc/systemd/system/backup-remote-to-local.timer
   register: backup_backup_remote_to_local_timer
-  changed_when: backup_backup_remote_to_local_timer.changed or activate_all_timers | bool
+  changed_when: backup_backup_remote_to_local_timer.changed or activate_all_timers | default(false) | bool
   notify: restart backup-remote-to-local timer
 
 - name: create backups-remote-to-local.sh

roles/cleanup-backups-timer/tasks/main.yml

@@ -3,7 +3,7 @@
     src:  "cleanup-backups.timer.j2"
     dest: "/etc/systemd/system/cleanup-backups.timer"
   register: cleanup_backups_timer
-  changed_when: cleanup_backups_timer.changed or activate_all_timers | bool
+  changed_when: cleanup_backups_timer.changed or activate_all_timers | default(false) | bool
   notify: restart cleanup-backups.timer
   when: run_once_cleanup_backups_timer is not defined
 

roles/cleanup-disc-space/tasks/main.yml

@@ -20,5 +20,5 @@
     src: cleanup-disc-space.timer.j2
     dest: /etc/systemd/system/cleanup-disc-space.timer
   register: cleanup_disc_space_timer
-  changed_when: cleanup_disc_space_timer.changed or activate_all_timers | bool
+  changed_when: cleanup_disc_space_timer.changed or activate_all_timers | default(false) | bool
   notify: restart cleanup-disc-space.timer

roles/docker-mailu/tasks/main.yml

@@ -56,5 +56,5 @@
     src:  "deploy-letsencrypt-mailu.timer.j2" 
     dest: "/etc/systemd/system/deploy-letsencrypt-mailu.timer"
   register: deploy_letsencrypt_mailu_timer
-  changed_when: deploy_letsencrypt_mailu_timer.changed or activate_all_timers | bool
+  changed_when: deploy_letsencrypt_mailu_timer.changed or activate_all_timers | default(false) | bool
   notify: restart deploy-letsencrypt-mailu.timer

roles/heal-docker/tasks/main.yml

@@ -23,7 +23,7 @@
     src: heal-docker.timer.j2
     dest: "/etc/systemd/system/heal-docker.timer"
   register: heal_docker_timer
-  changed_when: heal_docker_timer.changed or activate_all_timers | bool
+  changed_when: heal_docker_timer.changed or activate_all_timers | default(false) | bool
   notify: restart heal-docker.timer
   when: run_once_heal_docker is not defined
 

roles/health-btrfs/tasks/main.yml

@@ -20,5 +20,5 @@
     src: health-btrfs.timer.j2
     dest: "/etc/systemd/system/health-btrfs.timer"
   register: health_btrs_timer
-  changed_when: health_btrs_timer.changed or activate_all_timers | bool
+  changed_when: health_btrs_timer.changed or activate_all_timers | default(false) | bool
   notify: restart health-btrfs.timer

roles/health-disc-space/tasks/main.yml

@@ -20,5 +20,5 @@
     src: health-disc-space.timer.j2
     dest: /etc/systemd/system/health-disc-space.timer
   register: health_disc_space_timer
-  changed_when: health_disc_space_timer.changed or activate_all_timers | bool
+  changed_when: health_disc_space_timer.changed or activate_all_timers | default(false) | bool
   notify: restart health-disc-space.timer

roles/health-docker-container/tasks/main.yml

@@ -21,7 +21,7 @@
     src: health-docker-container.timer.j2
     dest: "/etc/systemd/system/health-docker-container.timer"
   register: health_docker_container_timer
-  changed_when: health_docker_container_timer.changed or activate_all_timers | bool
+  changed_when: health_docker_container_timer.changed or activate_all_timers | default(false) | bool
   notify: restart health-docker-container.timer
   when: run_once_health_docker_container is not defined
 

roles/health-docker-volumes/tasks/main.yml

@@ -21,7 +21,7 @@
     src: health-docker-volumes.timer.j2
     dest: "/etc/systemd/system/health-docker-volumes.timer"
   register: health_docker_volumes_timer
-  changed_when: health_docker_volumes_timer.changed or activate_all_timers | bool
+  changed_when: health_docker_volumes_timer.changed or activate_all_timers | default(false) | bool
   notify: restart health-docker-volumes.timer
   when: run_once_health_docker_volumes is not defined
 

roles/health-journalctl/tasks/main.yml

@@ -20,5 +20,5 @@
     src: health-journalctl.timer.j2
     dest: /etc/systemd/system/health-journalctl.timer
   register: health_journalctl_timer
-  changed_when: health_journalctl_timer.changed or activate_all_timers | bool
+  changed_when: health_journalctl_timer.changed or activate_all_timers | default(false) | bool
   notify: restart health-journalctl.timer

roles/health-nginx/tasks/main.yml

@@ -29,7 +29,7 @@
     src: health-nginx.timer.j2
     dest: "/etc/systemd/system/health-nginx.timer"
   register: health_nginx_timer
-  changed_when: health_nginx_timer.changed or activate_all_timers | bool
+  changed_when: health_nginx_timer.changed or activate_all_timers | default(false) | bool
   notify: restart health-nginx.timer
   when: run_once_health_nginx is not defined
 

roles/nginx-certbot/tasks/main.yml

@@ -16,7 +16,7 @@
     src:  certbot.timer.j2
     dest: /etc/systemd/system/certbot.timer
   register: certbot_timer
-  changed_when: certbot_timer.changed or activate_all_timers | bool
+  changed_when: certbot_timer.changed or activate_all_timers | default(false) | bool
   notify: restart certbot timer
   when: run_once_nginx_certbot is not defined
 

roles/system-maintenance-service-freezer/files/system-maintenance-service-freezer.py

@@ -4,14 +4,6 @@ import time
 import os
 from datetime import datetime
 
-# Global variable definition
-BREAK_TIME_SECONDS = 5
-FREEZER_SERVICES_PREFIX="system-maintenance-service-"
-
-class AttemptException(Exception):
-    """A custom exception for maximum number of attempts."""
-    pass
-
 def parse_time_to_seconds(time_str):
     """
     Convert a time string (e.g., '1h', '30min', '45s') to seconds.
@@ -55,36 +47,19 @@ def check_any_service_active(services):
     """
     return any(check_service_active(service) for service in services)
 
-def manage_timer(service, action):
-    """
-    Manage a systemd timer for a service.
-    action can be 'start' or 'stop'.
-    """
-    if action not in ['start', 'stop']:
-        raise ValueError("Invalid action specified for manage_timer")
-
-    timer_name = f"{service}.timer"
-    try:
-        subprocess.run(['systemctl', action, timer_name], check=True)
-        if action == 'start':
-            subprocess.run(['systemctl', 'enable', timer_name], check=True)
-        elif action == 'stop':
-            subprocess.run(['systemctl', 'disable', timer_name], check=True)
-        print(f"{timer_name} {action}ed and {'enabled' if action == 'start' else 'disabled'}.")
-    except subprocess.CalledProcessError as e:
-        print(f"Error managing timer {timer_name}: {e}")
-        exit(1)
-
 def stop_timer(service):
     """
     Stop and disable a systemd timer for a service if it exists.
     """
-    if service == f"{FREEZER_SERVICES_PREFIX}defrost":
+    if service == "system-maintenance-service-defrost":
         print(f"Ignoring {service}. It's the initializer of freezer.")
     if service_file_exists(service, "timer"):
-        manage_timer(service, 'stop')
+        timer_name = f"{service}.timer"
+        subprocess.run(['systemctl', 'stop', timer_name])
+        subprocess.run(['systemctl', 'disable', timer_name])
+        print(f"{timer_name} stopped and disabled.")
     else:
-        print(f"Timer {service}.timer does not exist.")
+        print("No timer to stop for service.")
 
 def filter_services(services, ignored_services):
     """
@@ -99,7 +74,7 @@ def stop_all_timers(services):
     for service in services:
         stop_timer(service)
 
-def wait_for_all_services_to_stop(filtered_services, max_attempts, attempt):
+def wait_for_all_services_to_stop(filtered_services, max_attempts, attempt, break_time_sec):
     """
     Wait until all services in the list have stopped, with a maximum number of attempts.
     """
@@ -107,45 +82,40 @@ def wait_for_all_services_to_stop(filtered_services, max_attempts, attempt):
         while check_service_active(service):
             attempt += 1
             if attempt > max_attempts:
-                raise AttemptException(f"Maximum attempts ({max_attempts}) reached. Exiting.")
+                raise Exception(f"Maximum attempts ({max_attempts}) reached. Exiting.")
-            print(f"{datetime.now().isoformat()}#{attempt}/{max_attempts}: Waiting for {BREAK_TIME_SECONDS} seconds for {service} to stop...")
+            print(f"{datetime.now().isoformat()}#{attempt}/{max_attempts}: Waiting for {break_time_sec} seconds for {service} to stop...")
-            time.sleep(BREAK_TIME_SECONDS)
+            time.sleep(break_time_sec)
     return attempt
 
 def freeze(filtered_services, timeout_sec):
     """
     Freeze services by stopping them and their timers, waiting up to a timeout.
     """
+    break_time_sec = 5
     attempt = 0
-    max_attempts = get_max_attempts(timeout_sec)
+    max_attempts = timeout_sec / break_time_sec
     
     while check_any_service_active(filtered_services):
         stop_all_timers(filtered_services)
-        attempt = wait_for_all_services_to_stop(filtered_services, max_attempts, attempt)
+        attempt = wait_for_all_services_to_stop(filtered_services, max_attempts, attempt, break_time_sec)
     print("All required services have stopped.")
 
-def get_max_attempts(timeout_sec):
-    return timeout_sec // BREAK_TIME_SECONDS
-
 def defrost(filtered_services,timeout_sec):
     """
     Defrost services by starting and enabling their timers.
     """
-    running_service = f"{FREEZER_SERVICES_PREFIX}defrost"
+    break_time_sec = 5
     attempt = 0
-    max_attempts = get_max_attempts(timeout_sec)
+    max_attempts = timeout_sec / break_time_sec
-    try:
+    wait_for_all_services_to_stop(filtered_services, max_attempts, attempt, break_time_sec)
-        wait_for_all_services_to_stop(filtered_services, max_attempts, attempt)
-    except AttemptException as e:
-        print(e)
-        print("Defrosting was not possible. The execution of other services took to long.")
-        manage_timer(running_service, "stop")
-        exit(0)
 
-    for service in filtered_services + [running_service]:
+    for service in filtered_services + [f"system-maintenance-service-{action}"]:
         print(f"Unfreezing: {service}")
         if service_file_exists(service, "timer"):
-            manage_timer(service, "start")
+            timer_name = f"{service}.timer"
+            subprocess.run(['systemctl', 'start', timer_name])
+            subprocess.run(['systemctl', 'enable', timer_name])
+            print(f"{timer_name} started and enabled.")
         else:
             print("No timer to activate for service.")
     print("All required services are started.")
@@ -156,7 +126,7 @@ def main(services, ignored_services, action, timeout_sec):
     """
     
     # Ignoring the current running service
-    running_service=f"{FREEZER_SERVICES_PREFIX}{action}"
+    running_service=f"system-maintenance-service-{action}"
     if running_service not in ignored_services:
         ignored_services.append(running_service)