当出现错误的daemon_reload=yes在守护进程重新加载时出现错误1:执行操作失败:连接超时

Question

处理程序的YAML文件如下所示: Yaml文件的位置如下

ansible_patching/ssm_agent.yml

---
- hosts: all
  gather_facts: true
  become: true

  roles:
  - ssm_agent

roles/ssm_agent/handlers/main.yml

# handlers file for ansible-role-ssm-agent
- name: systemctl_handler
  systemd:
    name: amazon-ssm-agent
    daemon_reload: yes
    state: "{{ ssm_agent_svc_state }}"
    enabled: yes

ansible_patching/roles/ssm_agent/tasks/main.yml

---
- name: Check required input parameters
  assert:
    that:
      - ({{ item }} is defined)
      - ({{ item }} is not none)
      - ({{ item }} | trim != '')
  with_items:
    - os_name
    - os_version
  loop_control:
          loop_var: item
- name: Include architecture specific
  include: "{{ ansible_architecture }}.yml"

- name: Include for centos 7
  include: install_{{ os_name }}_{{ os_version }}.yml
  when: ansible_distribution == "{{ os_name }}" and ansible_distribution_major_version == "{{ os_version }}"

我得到的错误如下

RUNNING HANDLER [ansible_role_ssm_agent : systemctl_handler] ********************************************************
fatal: [13.236.87.146]: FAILED! => {"changed": false, "msg": "failure 1 during daemon-reload: Failed to execute operation: Connection timed out\n"}

但是，我可以在远程CentOS主机的远程主机公共IP上使用相同的远程用户(Ansible)重新启动服务

请查阅以下产出：

[ansible@ip-172-31-33-3 tmp]$ sudo systemctl status amazon-ssm-agent
[sudo] password for ansible:
● amazon-ssm-agent.service - amazon-ssm-agent
   Loaded: loaded (/etc/systemd/system/amazon-ssm-agent.service; enabled; vendor preset: disabled)
   Active: active (running) since Thu 2020-06-25 01:22:40 UTC; 44min ago
 Main PID: 1978 (amazon-ssm-agen)
   CGroup: /system.slice/amazon-ssm-agent.service
           └─1978 /usr/bin/amazon-ssm-agent

[ansible@ip-172-31-33-3 tmp]$ sudo systemctl restart amazon-ssm-agent
[ansible@ip-172-31-33-3 tmp]$ sudo systemctl status amazon-ssm-agent
● amazon-ssm-agent.service - amazon-ssm-agent
   Loaded: loaded (/etc/systemd/system/amazon-ssm-agent.service; enabled; vendor preset: disabled)
   Active: active (running) since Thu 2020-06-25 02:07:06 UTC; 2s ago
 Main PID: 2157 (amazon-ssm-agen)
   CGroup: /system.slice/amazon-ssm-agent.service
           └─2157 /usr/bin/amazon-ssm-agent

注释：--如果我注释 daemon_reload: yes，那么剧本(ansible ssm代理安装角色)确实工作得很好

如有任何指导，将不胜感激。

[ansible@ip-172-31-38-88 ~]$ sudo systemctl daemon-reload
[ansible@ip-172-31-38-88 ~]$ echo $?
0

编辑2:使用sudo命令在远程主机上工作良好

[ansible@ip-172-31-38-88 ~]$ sudo cat /etc/sudoers |grep ansible
[sudo] password for ansible:
ansible ALL=(ALL)       ALL
[ansible@ip-172-31-38-88 ~]$

编辑4: Remote ansible用户具有与root用户相同的权限。

Answer 1

从您问题中的详细信息中可以看出，您正在使用ansible用户登录远程计算机。

为了启动-停止重新启动或重新加载系统服务，您的ansible用户必须拥有管理权限，即sudo权限。

从systemctl daemon-reload的输出中我们可以清楚地看到，您没有所需的权限。所以，为了让它发挥作用，

1. 您需要为ansible用户

授予sudo权限

在Ubuntu中，(使用root用户或sudo)：# usermod -aG sudo ansible

在CentOS中(与根用户或sudo一起)：# usermod -aG wheel ansible

1. 你需要在你的剧本中设置become: true，

---
- name: My playbook
  hosts: localhost
  become: true
  become_user: root
  become_method: sudo

  tasks:

仅此而已！