https://github.com/devture/com.devture.ansible.role.systemd_service_manager
An Ansible role which manages systemd services
https://github.com/devture/com.devture.ansible.role.systemd_service_manager
Last synced: 4 days ago
JSON representation
An Ansible role which manages systemd services
- Host: GitHub
- URL: https://github.com/devture/com.devture.ansible.role.systemd_service_manager
- Owner: devture
- Created: 2022-11-23T07:59:13.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2024-01-17T09:51:32.000Z (about 2 years ago)
- Last Synced: 2025-05-28T07:48:08.949Z (10 months ago)
- Language: Makefile
- Size: 14.6 KB
- Stars: 0
- Watchers: 2
- Forks: 2
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# systemd service manager Ansible role
This is an [Ansible](https://www.ansible.com/) role which manages systemd services.
## Features
- **starting** (restarting) services, in order, according to their `priority`. Services can all be stopped cleanly and then started anew, restarted one-by-one, or started in priority batches without blocking (see `devture_systemd_service_manager_service_restart_mode`)
- making services **auto-start** (see `devture_systemd_service_manager_services_autostart_enabled`)
- **verifying** services managed to start (see `devture_systemd_service_manager_up_verification_enabled`)
- **stopping** services, in order, according to their `priority`
- starting/stopping all defined services, or a group of services (`--tags=restart-group`, `--tags=stop-group`)
- restarting services by cleanly stopping them and restarting them, one by one, or by priority batches
## Usage
Example playbook:
```yaml
- hosts: servers
roles:
- when: devture_systemd_service_manager_enabled | bool
role: galaxy/com.devture.ansible.role.systemd_service_manager
```
Example playbook configuration (`group_vars/servers` or other):
```yaml
# See `devture_systemd_service_manager_services_list_auto` and `devture_systemd_service_manager_services_list_additional`
devture_systemd_service_manager_services_list_auto: |
{{
([{'name': 'some-service.service', 'priority': 1000}])
+
([{'name': 'another-service.service', 'priority': 1500}])
}}
```
Example playbook invocations tags (e.g. `ansible-playbook -i inventory/hosts setup.yml --tags=XXXXX`):
- `restart`, `restart-all`, `start-all` - restarts all services and potentially makes them auto-start (depending on `devture_systemd_service_manager_services_autostart_enabled`)
- `restart-group`, `start-group` - restarts services belonging to the specified group (e.g. `--extra-vars="group=core"`)
- `stop`, `stop-all` - stops all services
- `stop-group` - stops services belonging to the specified group (e.g. `--extra-vars="group=core"`)
Restart mode options for `devture_systemd_service_manager_service_restart_mode`:
- `all-at-once` (default) - issues a single `systemctl restart` (or `start`) command with all service names, letting systemd handle the entire transaction. See `defaults/main.yml` for detailed technical rationale
- `clean-stop-start` - stops all services in reverse priority order, then starts them in priority order
- `one-by-one` - restarts each service in priority order
- `priority-batched` - starts/restarts services in priority batches and queues them without blocking inside each batch
All-at-once retry controls:
- `devture_systemd_service_manager_all_at_once_retries` (default: `1`) - passed directly to task `retries` for each single all-at-once `systemctl restart ...` or `systemctl start ...` command; must be an integer `>= 1` (numeric strings accepted)
- `devture_systemd_service_manager_all_at_once_retry_delay_seconds` (default: `10`) - delay between attempts; must be a non-negative integer (numeric strings accepted)
For a detailed comparison of these modes with real-world downtime benchmarks, see [Restart Mode Comparison](docs/restart-mode-comparison.md).