我有一个由用户输入提供的服务器阵列,我想测试列表中的每个服务器,如果有任何故障,我想知道哪些服务器发生故障以及为什么(stderr)。请注意,服务器数量将不一致,因此该解决方案需要通过动态发送服务器数量。
命令
ansible-playbook test.yml -e '{"chrony_servers":["0.centos.pool.ntp.org","10.centos.pool.ntp.org"]}'
Test.yml
- name: Services Playbook
hosts: localhost
tasks:
- name: Install nmap-ncat
yum: name=nc state=present
- name: Validate chrony_servers
command: nc -vzu {{ item }} 123
with_items:
- "{{ chrony_servers }}"
ignore_errors: yes
register: chrony_server_results
- debug: var=chrony_server_results
- fail:
msg: "{{ chrony_server_results.results.stderr }}"
when: chrony_server_results.failed is defined and chrony_server_results.failed == true
这是此示例的输出
TASK [debug] ********************************************************************************************************************************************************************
ok: [localhost] => {
"chrony_server_results": {
"changed": true,
"failed": true,
"msg": "All items completed",
"results": [
{
"_ansible_ignore_errors": true,
"_ansible_item_label": "0.centos.pool.ntp.org",
"_ansible_item_result": true,
"_ansible_no_log": false,
"_ansible_parsed": true,
"changed": true,
"cmd": [
"nc",
"-vzu",
"0.centos.pool.ntp.org",
"123"
],
"delta": "0:00:02.217171",
"end": "2019-05-16 18:58:54.070600",
"failed": false,
"invocation": {
"module_args": {
"_raw_params": "nc -vzu 0.centos.pool.ntp.org 123",
"_uses_shell": false,
"argv": null,
"chdir": null,
"creates": null,
"executable": null,
"removes": null,
"stdin": null,
"warn": true
}
},
"item": "0.centos.pool.ntp.org",
"rc": 0,
"start": "2019-05-16 18:58:51.853429",
"stderr": "Ncat: Version 7.50 ( https://nmap.org/ncat )\nNcat: Connected to 208.75.88.4:123.\nNcat: UDP packet sent successfully\nNcat: 1 bytes sent, 0 bytes received in 2.21 seconds.",
"stderr_lines": [
"Ncat: Version 7.50 ( https://nmap.org/ncat )",
"Ncat: Connected to 208.75.88.4:123.",
"Ncat: UDP packet sent successfully",
"Ncat: 1 bytes sent, 0 bytes received in 2.21 seconds."
],
"stdout": "",
"stdout_lines": []
},
{
"_ansible_item_label": "10.centos.pool.ntp.org",
"_ansible_item_result": true,
"_ansible_no_log": false,
"_ansible_parsed": true,
"changed": true,
"cmd": [
"nc",
"-vzu",
"10.centos.pool.ntp.org",
"123"
],
"delta": "0:00:00.312970",
"end": "2019-05-16 18:58:54.510976",
"failed": true,
"invocation": {
"module_args": {
"_raw_params": "nc -vzu 10.centos.pool.ntp.org 123",
"_uses_shell": false,
"argv": null,
"chdir": null,
"creates": null,
"executable": null,
"removes": null,
"stdin": null,
"warn": true
}
},
"item": "10.centos.pool.ntp.org",
"msg": "non-zero return code",
"rc": 2,
"start": "2019-05-16 18:58:54.198006",
"stderr": "Ncat: Version 7.50 ( https://nmap.org/ncat )\nNcat: Could not resolve hostname \"10.centos.pool.ntp.org\": Name or service not known. QUITTING.",
"stderr_lines": [
"Ncat: Version 7.50 ( https://nmap.org/ncat )",
"Ncat: Could not resolve hostname \"10.centos.pool.ntp.org\": Name or service not known. QUITTING."
],
"stdout": "",
"stdout_lines": []
}
]
}
}
TASK [fail] *********************************************************************************************************************************************************************
fatal: [localhost]: FAILED! => {"msg": "The task includes an option with an undefined variable. The error was: 'list object' has no attribute 'stderr'\n\nThe error appears to have been in '/home/vera/test.yml': line 14, column 5, but may\nbe elsewhere in the file depending on the exact syntax problem.\n\nThe offending line appears to be:\n\n - debug: var=chrony_server_results\n - fail:\n ^ here\n"}
to retry, use: --limit @/home/vera/test.retry
PLAY RECAP **********************************************************************************************************************************************************************
预期结果应该类似于
"stderr": "Ncat: Version 7.50 ( https://nmap.org/ncat )\nNcat: Could not resolve hostname \"10.centos.pool.ntp.org\": Name or service not known. QUITTING."
答案 0 :(得分:2)
chrony_server_results
的结构是一个字典,其结构与调试任务假定的结构略有不同。
它在顶层具有一些键,这些键总体上适用于您正在调用with_items
的任务,例如changed
和failed
。
包含各自的stderr
和rc
属性的各个调用的“子结果”存储在具有键results
的数组中。
这意味着chrony_server_results.results.stderr
是错误的,并且该错误告诉您该数组没有stderr
属性,而单个条目有多个属性,您需要对其进行迭代。
您可以检查rc
来了解相应的调用是否真的 失败了(因为netcat显然在所有情况下都写入stderr)。
以下解决方案首先构建一个数组,该数组仅包含rc
大于零的结果,然后仅显示针对那些结果的调试消息:
- set_fact:
failed_results: "{{ chrony_server_results.results | selectattr('rc', '>', 0) | list }}"
- name: Show the stderrs of failed_results
debug:
msg: "{{ item.stderr }}"
with_items: "{{ failed_results }}"
- fail:
msg: "See debug output above"
when: failed_results | length > 0
第一个表达式中使用的过滤器是jinja2的一部分,并记录在这里:http://jinja.pocoo.org/docs/2.10/templates/#selectattr