Ansible扁平配置文件：能否用分段配置替代子目录及多文件变量？

Absolutely, you can absolutely do both of these—Ansible is flexible enough to support consolidated file structures alongside its traditional directory-based approach. Let’s break this down for each scenario:

Yes, this is fully supported. Ansible’s role system doesn’t force you to split tasks, handlers, vars, etc., into separate subdirectories. You can bundle all role components into a single roles/[role_name]/main.yml file by using YAML’s top-level keys to segment different parts of the role.

For example, your roles/web_server/main.yml might look like this:

---
# Tasks section
tasks:
  - name: Install Nginx
    apt:
      name: nginx
      state: present

  - name: Enable Nginx service
    service:
      name: nginx
      enabled: true

# Handlers section
handlers:
  - name: Restart Nginx
    service:
      name: nginx
      state: restarted

# Variables section
vars:
  nginx_listen_port: 80
  nginx_server_name: "example.com"

# Defaults section (lower priority than vars)
defaults:
  nginx_max_clients: 1024

This works perfectly for smaller, simpler roles. For large, complex roles with dozens of tasks or handlers, the directory structure (with separate tasks/, handlers/, etc., folders) might still be easier to navigate—but the single-file approach is a valid alternative.

You can absolutely consolidate variables for multiple groups or hosts into a single file instead of creating individual files per group/host. Here’s how:

For group_vars #

Instead of creating group_vars/web_servers.yml, group_vars/db_servers.yml, etc., you can create a single file (e.g., group_vars/consolidated.yml) and structure it as a YAML dictionary where each key is the group name, and the value is the set of variables for that group:

---
web_servers:
  nginx_port: 80
  ssl_enabled: true
  app_version: "1.2.3"

db_servers:
  db_engine: postgresql
  db_name: app_database
  db_user: app_user

all:
  global_timeout: 300
  backup_server: "backup.example.com"

Ansible will automatically map these variables to the corresponding groups when running playbooks. Note that if you have both a consolidated file and individual group files (like group_vars/web_servers.yml), the individual files will take precedence (higher priority) over the consolidated ones.

For host_vars #

Similarly, you can consolidate host-specific variables into a single file (e.g., host_vars/all_hosts.yml) using a dictionary structure:

---
web01.example.com:
  nginx_port: 8080
  custom_log_path: "/var/log/nginx/web01.log"

db01.example.com:
  db_memory_allocation: "4G"
  backup_frequency: "daily"

Tradeoffs to consider #

• Pros: Fewer files to manage, easier to get a high-level view of all variables in one place.
• Cons: As your inventory grows (more groups/hosts), the single file can become bloated and harder to scan for specific variables. For large environments, individual files might still be more maintainable.

内容的提问来源于stack exchange，提问作者Danny F