k8s_cp inconsistent behavior with parameter local_path

[haroonb]

SUMMARY

ISSUE TYPE

• Bug Report

COMPONENT NAME

ANSIBLE VERSION

ansible [core 2.12.2]
  config file = /etc/ansible/ansible.cfg
  configured module search path = ['/home/butt/.ansible/plugins/modules', '/usr/share/ansible/plugins/modules']
  ansible python module location = /usr/lib/python3/dist-packages/ansible
  ansible collection location = /home/butt/.ansible/collections:/usr/share/ansible/collections
  executable location = /usr/bin/ansible
  python version = 3.8.10 (default, Nov 26 2021, 20:14:08) [GCC 9.3.0]
  jinja version = 2.10.1
  libyaml = True

COLLECTION VERSION

# /usr/share/ansible/collections/ansible_collections
Collection           Version
-------------------- -------
amazon.aws           3.1.1
ansible.netcommon    2.0.1
ansible.posix        1.3.0
ansible.utils        2.0.2
cloud.common         2.1.0
community.aws        3.1.0
community.crypto     1.7.1
community.docker     2.2.1
community.general    4.6.1
community.grafana    1.3.3
community.kubernetes 2.0.1
community.postgresql 2.1.2
community.vmware     1.0.0
google.cloud         1.0.2
kubernetes.core      2.3.0

CONFIGURATION

CALLBACKS_ENABLED(/etc/ansible/ansible.cfg) = ['profile_tasks']
DEFAULT_JINJA2_NATIVE(/etc/ansible/ansible.cfg) = True
DEFAULT_LOG_PATH(/etc/ansible/ansible.cfg) = /var/log/ansible.log
HOST_KEY_CHECKING(/etc/ansible/ansible.cfg) = False

OS / ENVIRONMENT

STEPS TO REPRODUCE

You can repoduce this with the following steps:

• copy from_pod to local_path
• copy to_pod from same local_path

---
- hosts: devops-m1
  gather_facts: no
  tasks:

    - name: create busy box
      kubernetes.core.k8s:
        wait: yes
        state: present
        definition:
          apiVersion: v1
          kind: Pod
          metadata:
            name: concept
            namespace: default
          spec:
            containers:
            - name: busybox
              image: busybox
              args:
              - sleep
              - infinity

    - name: create a file
      kubernetes.core.k8s_exec:
        pod: concept
        namespace: default
        command: touch /tmp/test.file

    - name: copy from_pod to local_path
      kubernetes.core.k8s_cp:
        state: from_pod
        pod: concept
        namespace: default
        remote_path: /tmp/test.file
        local_path: /tmp/test.file

    - name: check if file exists
      stat:
        path: /tmp/test.file
        get_checksum: yes
      register: result

    - name: copy to_pod from local_path
      kubernetes.core.k8s_cp:
        state: to_pod
        pod: concept
        namespace: default
        remote_path: /tmp/file.test
        local_path: /tmp/test.file
      when: result.stat.exists

EXPECTED RESULTS

I expected the module to use the same local_path on the remote host to read the file from, which it used to write the files to.

The documentation of the module.

# kubectl cp /tmp/foo some-namespace/some-pod:/tmp/bar -c some-container
- name: Copy /tmp/foo local file to /tmp/bar in a remote pod in a specific container
  kubernetes.core.k8s_cp:
    namespace: some-namespace
    pod: some-pod
    container: some-container
    remote_path: /tmp/bar
    local_path: /tmp/foo
    no_preserve: True
    state: to_pod

ACTUAL RESULTS

The modules tries to copy from the ansible controller

The full traceback is:
Traceback (most recent call last):
  File "/usr/share/ansible/collections/ansible_collections/kubernetes/core/plugins/action/k8s_cp.py", line 271, in get_file_realpath
    return self._find_needle("files", local_path)
  File "/usr/lib/python3/dist-packages/ansible/plugins/action/__init__.py", line 1364, in _find_needle
    return self._loader.path_dwim_relative_stack(path_stack, dirname, needle)
  File "/usr/lib/python3/dist-packages/ansible/parsing/dataloader.py", line 341, in path_dwim_relative_stack
    raise AnsibleFileNotFound(file_name=source, paths=[to_native(p) for p in search])
ansible.errors.AnsibleFileNotFound: Could not find or access '/tmp/test.file' on the Ansible Controller.
If you are using a module and expect the file to exist on the remote, see the remote_src option
fatal: [devops-m1]: FAILED! => {
    "changed": false,
    "msg": "/tmp/test.file does not exist in local filesystem"
}

[abikouo]

@haroonb Thanks for submitting this issue.
Your playbook delegates task to your host devops-m1, while trying to copy to_pod the action plugin find the file on the controller node (localhost) not on the managed node (devops-m1) (extract from documentation Action plugins let you integrate local processing and local data with module functionality. https://docs.ansible.com/ansible/latest/dev_guide/developing_plugins.html), this is why we are getting such error.
We need to add a specific parameter such as remote=yes to state that the file is located on the managed node.

[abikouo]

@haroonb could you please test using the PR #422 ? Thanks

[haroonb]

Hello @abikouo, I made the change in my local collection and can confirm, that it works.

[abikouo]

Hello @abikouo, I made the change in my local collection and can confirm, that it works.

Thanks for the feedback!!