Ansible Playbook主机不可达场景下触发block的rescue块执行问题

你可以通过「先忽略不可达错误、再手动判断触发失败」的逻辑实现需求，既不会让playbook全局终止，也能正常触发rescue块。

实现逻辑 #

• 给启动Apache的任务添加ignore_unreachable: yes参数，避免主机不可达时直接终止playbook，同时注册任务执行结果变量
• 在Apache启动任务后新增失败断言任务，若检测到主机不可达则主动抛出失败，触发rescue逻辑
• 主机可达时断言任务自动跳过，不影响原有场景1、2的原生block/rescue执行逻辑

调整后的Playbook示例 #

tasks:
  - name: "Task 1"
  ...

  - name: "Start Apache"
    block:
    - name: "Ensure Apache is started"
      ansible.builtin.service:
          name: apache2
          state: started
      ignore_unreachable: yes
      register: apache_start_result

    - name: "判断主机是否不可达"
      ansible.builtin.fail:
        msg: "目标主机 {{ inventory_hostname }} SSH连接失败"
      when: apache_start_result.unreachable is defined and apache_start_result.unreachable

    - name: "Wait for website to come up on this host"
      delegate_to: 127.0.0.1
      uri:
        url: "http://{{inventory_hostname}}"
        return_content: yes
      register: _result
      until: _result.content != '\"\"'
      retries: 5
      delay: 5
  
    rescue:
    - name: "Do Something"
      ...

内容的提问来源于stack exchange，提问作者john smith