播放过程中出现SSH错误

我在Ansible上犯了一个奇怪的错误。首先，第一个角色工作正常，但是当Ansible尝试执行第二个角色时，由于ssh错误，它失败了

环境：

ansible-playbook -vvvv -i inventory/dev playbook_update_system.yml --limit "db[0]"

- name: "HUB Playbook | Updating system packages on {{ ansible_hostname }}"
  hosts: release_first_half
  roles:
    - upgrade_system_package
    - reboot_server

- name: "upgrading CentOS system packages on {{ ansible_hostname }}"
  shell: sudo puppet apply -e 'exec{"upgrade-package":command => "/usr/bin/yum clean all; /usr/bin/yum -y update;"}'
  when: ansible_distribution == 'CentOS' and 'cassandra' not in group_names

- name: "reboot CentOS [{{ ansible_hostname }}] server"
  shell: sudo puppet apply -e 'exec{"reboot-os":command => "/usr/sbin/reboot"}'
  when: ansible_distribution == 'CentOS' and 'cassandra' not in group_names

• 操作系统：CentOS 7
• Ansible版本：2.2.1.0
• Python版本：2.7.5
• OpenSSH版本：OpenSSH_6.6.1p1、OpenSSL 1.0.1e-fips 2013年2月11日

执行的Ansible命令：

ansible-playbook -vvvv -i inventory/dev playbook_update_system.yml --limit "db[0]"

- name: "HUB Playbook | Updating system packages on {{ ansible_hostname }}"
  hosts: release_first_half
  roles:
    - upgrade_system_package
    - reboot_server

- name: "upgrading CentOS system packages on {{ ansible_hostname }}"
  shell: sudo puppet apply -e 'exec{"upgrade-package":command => "/usr/bin/yum clean all; /usr/bin/yum -y update;"}'
  when: ansible_distribution == 'CentOS' and 'cassandra' not in group_names

- name: "reboot CentOS [{{ ansible_hostname }}] server"
  shell: sudo puppet apply -e 'exec{"reboot-os":command => "/usr/sbin/reboot"}'
  when: ansible_distribution == 'CentOS' and 'cassandra' not in group_names

剧本：

ansible-playbook -vvvv -i inventory/dev playbook_update_system.yml --limit "db[0]"

- name: "HUB Playbook | Updating system packages on {{ ansible_hostname }}"
  hosts: release_first_half
  roles:
    - upgrade_system_package
    - reboot_server

- name: "upgrading CentOS system packages on {{ ansible_hostname }}"
  shell: sudo puppet apply -e 'exec{"upgrade-package":command => "/usr/bin/yum clean all; /usr/bin/yum -y update;"}'
  when: ansible_distribution == 'CentOS' and 'cassandra' not in group_names

- name: "reboot CentOS [{{ ansible_hostname }}] server"
  shell: sudo puppet apply -e 'exec{"reboot-os":command => "/usr/sbin/reboot"}'
  when: ansible_distribution == 'CentOS' and 'cassandra' not in group_names

角色：升级系统软件包：

ansible-playbook -vvvv -i inventory/dev playbook_update_system.yml --limit "db[0]"

- name: "HUB Playbook | Updating system packages on {{ ansible_hostname }}"
  hosts: release_first_half
  roles:
    - upgrade_system_package
    - reboot_server

- name: "upgrading CentOS system packages on {{ ansible_hostname }}"
  shell: sudo puppet apply -e 'exec{"upgrade-package":command => "/usr/bin/yum clean all; /usr/bin/yum -y update;"}'
  when: ansible_distribution == 'CentOS' and 'cassandra' not in group_names

- name: "reboot CentOS [{{ ansible_hostname }}] server"
  shell: sudo puppet apply -e 'exec{"reboot-os":command => "/usr/sbin/reboot"}'
  when: ansible_distribution == 'CentOS' and 'cassandra' not in group_names

角色：重新启动服务器：

ansible-playbook -vvvv -i inventory/dev playbook_update_system.yml --limit "db[0]"

- name: "HUB Playbook | Updating system packages on {{ ansible_hostname }}"
  hosts: release_first_half
  roles:
    - upgrade_system_package
    - reboot_server

- name: "upgrading CentOS system packages on {{ ansible_hostname }}"
  shell: sudo puppet apply -e 'exec{"upgrade-package":command => "/usr/bin/yum clean all; /usr/bin/yum -y update;"}'
  when: ansible_distribution == 'CentOS' and 'cassandra' not in group_names

- name: "reboot CentOS [{{ ansible_hostname }}] server"
  shell: sudo puppet apply -e 'exec{"reboot-os":command => "/usr/sbin/reboot"}'
  when: ansible_distribution == 'CentOS' and 'cassandra' not in group_names

当前行为：

ansible-playbook -vvvv -i inventory/dev playbook_update_system.yml --limit "db[0]"

- name: "HUB Playbook | Updating system packages on {{ ansible_hostname }}"
  hosts: release_first_half
  roles:
    - upgrade_system_package
    - reboot_server

- name: "upgrading CentOS system packages on {{ ansible_hostname }}"
  shell: sudo puppet apply -e 'exec{"upgrade-package":command => "/usr/bin/yum clean all; /usr/bin/yum -y update;"}'
  when: ansible_distribution == 'CentOS' and 'cassandra' not in group_names

- name: "reboot CentOS [{{ ansible_hostname }}] server"
  shell: sudo puppet apply -e 'exec{"reboot-os":command => "/usr/sbin/reboot"}'
  when: ansible_distribution == 'CentOS' and 'cassandra' not in group_names

连接到“db1”节点并执行角色“升级系统包”=>OK

尝试连接到“db1”并执行角色“reboot_server”=>由于ssh失败

Ansible返回的错误消息：

ansible-playbook -vvvv -i inventory/dev playbook_update_system.yml --limit "db[0]"

- name: "HUB Playbook | Updating system packages on {{ ansible_hostname }}"
  hosts: release_first_half
  roles:
    - upgrade_system_package
    - reboot_server

- name: "upgrading CentOS system packages on {{ ansible_hostname }}"
  shell: sudo puppet apply -e 'exec{"upgrade-package":command => "/usr/bin/yum clean all; /usr/bin/yum -y update;"}'
  when: ansible_distribution == 'CentOS' and 'cassandra' not in group_names

- name: "reboot CentOS [{{ ansible_hostname }}] server"
  shell: sudo puppet apply -e 'exec{"reboot-os":command => "/usr/sbin/reboot"}'
  when: ansible_distribution == 'CentOS' and 'cassandra' not in group_names

致命：[db1]：无法访问！=>{ “更改”：错误， “味精”："无法通过ssh连接到主机：OpenSSH_6.6.1，OpenSSL 1.0.1e-fips 2013年2月11日\r\ndebug1:读取配置数据/USR/newtprod/.ssh/config\r\ndebug1:读取配置数据/etc/ssh/ssh_-config\r\ndebug1:/etc/ssh/ssh-config第56行：应用*\r\ndebug1:自动mux:尝试现有主\r\ndebug2:fd 3设置O\r\n块\r\ndebug2:mux\u客户端\u hello\u交换：主版本4\r\ndebug3:mux\u客户端\u转发：请求转发：0本地，0远程\r\ndebug3:mux\u客户端\u请求\u会话：输入\r\ndebug3:mux\u客户端\u请求\u活动：完成pid=64994\r\ndebug3:mux\u客户端\u请求\u会话：发送的会话请求\r\ndebug1:mux_客户端\u请求\u会话：主会话id:2\r\n错误3:mux\u客户端\u读取\u数据包：读取头失败：管道断开\r\n错误2:Control master意外终止\r\n与db1的共享连接已关闭。\r\n“， “遥不可及”：真实 }

---

我不明白，因为上一个角色已经在此节点上成功执行。此外，我们有很多playbook使用相同的清单文件，它们工作正常。我也在另一个节点上尝试过，但结果相同。

这是一个简单且众所周知的问题：关闭过程会导致SSH守护进程退出，从而中断current SSH会话（出现“断管”错误）。服务器正确重新启动，但Ansible流中断

您需要向

shell

命令添加延迟，并使用

async

选项运行它，以便Ansible的SSH会话可以在终止之前完成

shell: sleep 5; sudo puppet apply -e 'exec{"reboot-os":command => "/usr/sbin/reboot"}'
async: 0
poll: 0