英文:
How to re-run a task if its result fails
问题
---
- name: 获取md5
net_command:
commands:
- "bash md5sum {{ image }}"
register: md5_remote
- name: 验证md5
assert:
that:
- md5_local in md5_remote.stdout[0]
register: md5_result
until: "md5_result is not failed"
retries: 5
delay: 60
请注意，net_command 是我们用于某些特定网络设备的自定义模块，没有内置的 Ansible 模块。
因此，我尝试的是，如果 Verify md5 断言失败，重新运行 Get md5 任务。目前，它只是在 Verify md5 上循环，而实际上没有重新计算检验和，所以如果它失败一次，它将永远无法完成。
所以，基本上，我需要多次重新计算检验和，直到它与存储在 md5_local 中的预期值匹配。
英文:
I have a couple of tasks like this:
---
- name: Get md5
net_command:
commands:
- "bash md5sum {{ image }}"
register: md5_remote
- name: Verify md5
assert:
that:
- md5_local in md5_remote.stdout[0]
register: md5_result
until: "md5_result is not failed"
retries: 5
delay: 60
Note that net_command is a custom module we use for some specific networking equipment that does not have any builtin Ansible modules.
So what I am trying to do is to re-run the Get md5 task if the Verify md5 assertion fails. Right now it's just looping over Verify md5 without actually re-calculating the checksum again, so, if it fails once it will never work complete.
So, basically, I need to recaculate the checksum multiple times until it matches the expected one, stored in md5_local.
答案1
得分: 2
如果您想重试一个任务，您需要在该任务上使用until循环。由于until已经是一个断言，所以不需要单独的assert任务。
因此，您的用例应该可以通过单个任务来实现：
- name: 获取 md5
net_command:
commands:
- "bash md5sum {{ image }}"
register: md5_remote
until: md5_local in md5_remote.stdout[0]
retries: 5
delay: 60
英文:
If you want to retry a task, you'll have to put the until loop on that task. Since until is an assertion already, there is no need for a separate assert task.
So, your use case should be covered by a single task:
- name: Get md5
net_command:
commands:
- "bash md5sum {{ image }}"
register: md5_remote
until: md5_local in md5_remote.stdout[0]
retries: 5
delay: 60
通过集体智慧和协作来改善编程学习和解决问题的方式。致力于成为全球开发者共同参与的知识库，让每个人都能够通过互相帮助和分享经验来进步。
评论